Extending Full-Plate Tectonic Models into Deep Time: Linking the and the

Andrew S. Merdith1,*, Simon E. Williams2, Alan S. Collins3, Michael G. Tetley1, Jacob A. Mulder4, Morgan L. Blades3, Alexander Young5, Sheree E. Armistead6, John Cannon7, Sabin Zahirovic7 and R. Dietmar Müller7 1 UnivLyon, Université Lyon 1, Ens de Lyon, CNRS, UMR 5276 LGL-TPE, F-69622, Villeurbanne, France 2 Northwest University, Xi’an, China 3 and Systems (TES) Group, Departtment of Earth Sciences, The University of Adelaide, Adelaide, SA 5005, 4 School of Earth, Atmosphere and Environment, Monash University, Clayton, Victoria 3168, Australia 5 GeoQuEST Research Centre, School of Earth, Atmospheric and Life Sciences, University of Wollongong, Northfields Avenue, NSW 2522, Australia 6 Geological Survey of , 601 Booth Street, Ottawa, Ontario, Canada & Metal Earth, Harquail School of Earth Sciences, Laurentian University, Sudbury, Ontario, Canada 7 Earthbyte Group, School of Geosciences, University of Sydney, Sydney, , 2006, Australia [email protected] @AndrewMerdith [email protected] [email protected] @geoAlanC [email protected] @mikegtet [email protected] [email protected] [email protected] [email protected] @geoSheree [email protected] [email protected] @tectonicSZ [email protected] @MullerDietmar

This is the unformatted accepted manuscript published in Earth-Science Reviews https://www.journals.elsevier.com/earth-science-reviews

DOI: https://doi.org/10.1016/j.earscirev.2020.103477

1 Extending Full-Plate Tectonic Models into Deep Time: Linking the Neoproterozoic and the 2 Phanerozoic 3 4 Andrew S. Merdith1,*, Simon E. Williams2, Alan S. Collins3, Michael G. Tetley1, Jacob A. Mulder4, Morgan 5 L. Blades3, Alexander Young5, Sheree E. Armistead6, John Cannon7, Sabin Zahirovic7 and R. Dietmar 6 Müller7 7 1 UnivLyon, Université Lyon 1, Ens de Lyon, CNRS, UMR 5276 LGL-TPE, F-69622, Villeurbanne, France 8 2 Northwest University, Xi’an, China 9 3 Tectonics and Earth Systems (TES) Group, Departtment of Earth Sciences, The University of Adelaide, Adelaide, SA 5005, Australia 10 4 School of Earth, Atmosphere and Environment, Monash University, Clayton, Victoria 3168, Australia 11 5 GeoQuEST Research Centre, School of Earth, Atmospheric and Life Sciences, University of Wollongong, Northfields Avenue, NSW 2522, 12 Australia 13 6 Geological Survey of Canada, 601 Booth Street, Ottawa, Ontario, Canada & Metal Earth, Harquail School of Earth Sciences, Laurentian 14 University, Sudbury, Ontario, Canada 15 7 Earthbyte Group, School of Geosciences, University of Sydney, Sydney, New South Wales, 2006, Australia 16 * Corresponding author: [email protected] 17 18 Abstract 19 20 Recent progress in plate tectonic reconstructions has seen models move beyond the classical idea of 21 continental drift by attempting to reconstruct the full evolving configuration of tectonic plates and plate 22 boundaries. A particular problem for the Neoproterozoic and is that many existing interpretations 23 of geological and palaeomagnetic data have remained disconnected from younger, better-constrained 24 periods in Earth history. An important test of deep time reconstructions is therefore to demonstrate the 25 continuous kinematic viability of tectonic motions across multiple cycles. We present, for 26 the first time, a continuous full-plate model spanning 1 Ga to the present-day, that includes a revised and 27 improved model for the Neoproterozoic–Cambrian (1000–520 Ma) that connects with models of the 28 Phanerozoic, thereby opening up pre- times for quantitative analysis and further regional 29 refinements. In this contribution, we first summarise methodological approaches to full-plate modelling 30 and review the existing full-plate models in order to select appropriate models that produce a single 31 continuous model. Our model is presented in a palaeomagnetic reference frame, with a newly-derived 32 apparent polar wander path for Gondwana from 540 to 320 Ma, and a global apparent polar wander path 33 from 320 to 0 Ma. We stress, though while we have used palaeomagnetic data when available, the model 34 is also geologically constrained, based on preserved data from past-plate boundaries. This study is intended 35 as a first step in the direction of a detailed and self-consistent tectonic reconstruction for the last billion 36 of Earth history, and our model files are released to facilitate community development. 37 38 1 Introduction 39 40 is a unifying theory of modern , explicitly connecting the and processes 41 that bridge the , , and atmosphere. Tectonic forces control the rates of uplift 42 and erosion where collide or separate (England and Molnar, 1990) and modulate the flow of 43 energy between , lithosphere and mantle as continental configurations evolve (Bebout, 1995; Karlsen 44 et al., 2019; Müller et al., 2008). Evolving plate tectonic configurations also determine changes in how 45 species are distributed across different landmasses (McKenzie et al., 2014; Meert and Lieberman, 2008) 46 and infer the rates of chemical flux between the Earth’s surface and the deep interior (Gernon et al., 2016; 47 Jarrard, 2003). 48 49 Global reconstructions have traditionally focussed on the positions of the major continents and geological 50 preserved within them. Data acquired from modern oceans provide a powerful constraint on the 51 breakup of the supercontinent Pangea over the last ca. 200 Ma, and form the basis of continuous models of 52 plate configurations from the to present (e.g. Müller et al., 2016; Seton et al., 2012). These ‘full- 53 plate’ reconstructions use geological and geophysical data to determine the configurations and motions of 54 both continental and oceanic lithosphere, and the of the plate boundaries that separate neighbouring 55 plates. Together with the development of free software tools (Boyden et al., 2011; Müller et al., 2018), full- 56 plate reconstructions permit quantitative estimates of tectonic processes through time within a continuous, 57 consistent kinematic framework, opening up portions of Earth’s history to quantitative analysis (e.g. Bower 58 et al., 2013; Brune et al., 2017; Dutkiewicz et al., 2019; Hounslow et al., 2018; Karlsen et al., 2019; Merdith 59 et al., 2019a). 60 61 Plate tectonic processes are thought to have been the dominant control on Earth’s paleogeography possibly 62 since 3.2 Ga (Brenner et al., 2020; Brown et al., 2020a; Cawood et al., 2018a; Gerya, 2014; Palin et al., 63 2020). Studies of the pre-Pangean Earth have led to the proposal that Pangea was preceded by the 64 (Dalziel, 1991; Hoffman, 1991; Moores, 1991) and Nuna/Columbia 65 (Meert, 2002; Rogers and Santosh, 2002; Zhao et al., 2002) and earlier Archaean ‘supercratons’ (e.g. 66 Bleeker, 2003; Pehrsson et al., 2013; Smirnov et al., 2013), reflecting transient aggregations of continental 67 blocks interspersed between other phases of Earth’s history when the continents were more dispersed. The 68 absence of a pre-Mesozoic floor record neccessitates that reconstructing the pre-Pangean Earth relies 69 on the fragmented geological record preserved within the continents. Early studies of Proterozoic 70 supercontinents provide individual snapshots of continental configurations; though there are differences 71 between competing interpretations. More recently, attempts have been made to reconcile Neoproterozoic 72 continental motions within a continuous kinematic framework (Cawood et al., 2020; Collins and 73 Pisarevsky, 2005; Li et al., 2008). To further infer the extent and nature of tectonic boundaries covering all 74 of Earth’s surface in the Proterozoic requires methodical extrapolation of available observations and is 75 subject to major uncertainties. Despite this, these reconstructions are valuable in that they make testable 76 predictions about and time periods where observations are lacking. 77 78 Full-plate models published over the last decade collectively span the last 1 Ga. However, each of these 79 models cover different time periods or areas of the and each model is based on different assumptions 80 and hypotheses, and place differing emphases on subsets of the geological record. Thus, although 81 continental motions and plate boundary evolution have been categorised in some manner for the past 1 Ga, 82 there is no fully continuous model defining Earth’s tectonic history for this time. A fundamental test of any 83 tectonic reconstruction for the is that the configurations of continents, terranes and plate 84 boundaries can evolve continuously as to seamlessly merge with reconstructed configurations for more 85 recent times that are better constrained and ultimately tied to the present-day Earth. The absence of such 86 continuous reconstructions highlights a critical uncertainty for assessing interpretations of Neoproterozoic 87 , tectonics and geodynamics. 88 89 Our key motivations for this study are three-. Firstly, a 1 Ga model will permit, for the first time, 90 Neoproterozoic and Cambrian quantitative analysis that constrains (bio)geochemical and volatile fluxes, 91 palaeoclimatic studies and the nature of earth systems, during times of biological evolution and extreme 92 climate change (Gernon et al., 2016; Goddéris et al., 2017; Mills et al., 2011, 2019). Second, a full-plate 93 model would be a starting point for future studies to constrain both the tectonic (e.g. 94 (Li et al., 2019; Merdith et al., 2019b)) and geodynamic (e.g. core-lithosphere-mantle connection (Heron et 95 al., 2020; Tetley et al., 2019)) nature and evolution of the Earth. Third, a consistent model for the 96 Neoproterozoic and Cambrian that coherently links with younger models can be used as a framework to 97 support future regional studies that test and enhance the resolution of the model or spawn alternative models 98 that can be used for hypothesis testing. We stress that our reconstruction is intended to capture the main 99 aspects of global tectonics across the last billion years and consequently lacks many details that could be 100 incorporated for individual regions. Just as the earliest full-plate models for the and Mesozoic 101 (Seton et al., 2012) and late Palaeozoic (Domeier and Torsvik, 2014) have provided valuable open-access 102 resources for numerous other studies to test and improve, we intend that the global framework provided by 103 our reconstruction will form the basis for future studies that will generate improved reconstructions by 104 incorporating new or different observations and ideas. 105 106 This paper is organised as follows: first, we provide a review of the concepts behind full-plate 107 reconstructions, including the types of observations and assumptions on which they are based. We then 108 summarise the previously published reconstruction models and justify which elements of these existing 109 studies we have chosen to include in our reconstruction. Finally, we present the main outcome of our study: 110 the first continuous and self-consistent full-plate model from 1 Ga to present day with a single set of 111 polygons, Euler rotations and plate boundaries. 112 113 2 Full- models 114 There are two broad categories of models that can be constructed to describe Earth’s tectonic or 115 palaeogeographic history. The first category we refer to as ‘continental drift’ type models (Fig. 1a), as they 116 model the motion of continents drifting across the Earth’s surface and tend to explicitly reflect 117 palaeogeography rather than tectonic evolution. The second type we refer to as ‘full-plate models’ (Fig. 118 1b), which, in addition to tracking the motion of continents, trace the evolution of plate boundaries and by 119 implication, the evolution of tectonic plates themselves (Gurnis et al., 2012). In effect, continental drift type 120 models are the precursor to full-plate models, but rather than supersede continental drift models, both types 121 of models complement each other and provide different avenues for research. Continental drift type models 122 are useful for analysing palaeomagnetic data, for contextualising regional studies or as a ground-breaking 123 study where there is little preserved data on plate boundaries. Comparably, full-plate models are more 124 encompassing, but are also much harder to iterate over and generate alternative models from. If one simply 125 requires the distribution of continental and not of plate boundaries, then it is much easier and simpler 126 to build a continental drift style model than categorically describe and model plate boundaries through time. 127 However, both types of models use the same reconstruction framework. 128 129

Figure 1 Schematic comparison of evolution of plate tectonic modelling. (a) ‘Continental drift’/palaeogeographic type models and (b) full-plate models; (i)–(iv) identify separate plates. Palaeomagnetic data are the primary constraint of the movement of continents in both (a) and (b) however, the inclusion of geological data into the model in (b) preserves the relative type of motion between two continents (divergence, convergence or transform) and allows for the construction of plate boundaries. 130 131 2.1 Reconstruction Framework 132 133 The essential characteristic of any plate reconstruction is the reconstruction framework (or network 134 (Domeier and Torsvik, 2017)), which is the organisation of data used to describe the motion of rigid objects 135 on a sphere using Euler’s rotation theorem (McKenzie and Parker, 1967; Morgan, 1968). In plate 136 reconstructions, the rigid objects in motion are the plates themselves and, in addition to requiring the 137 temporal and spatial components of moving plates (i.e. the time period of motion, latitude, longitude and 138 angle of rotation) the theorem also requires that each rotation be defined relative to another object. This 139 rotation of one object relative to another forms the basis of a ‘relative plate motion model’. It is also 140 desirable for these relative rotations to be tied to something (relatively) immutable within, or around, the 141 Earth (e.g. the core, mantle, or the spin-axis) thus, transforming the model into an ‘absolute plate motion 142 model’. For the Mesozoic and Cenozoic, plates are described in a relative framework due to the preservation 143 of oceanic lithosphere within oceans formed since the breakup of Pangea. In this way each plate’s motion 144 history is described as moving relative to another plate, with the typically at the top of the 145 hierarchy (e.g. (Ross and Scotese, 1988; Torsvik et al., 2008); Fig. 2a, b). The motion of the African plate 146 can then be defined absolutely (though using observations from many or all plates, and not just ) to 147 the deep Earth through alternative methods such as chains (Müller et al., 1993; O’Neill et al., 2005), 148 seismic imaging of subducted slabs (van der Meer et al., 2010), palaeomagnetic data (with or without true 149 polar wander corrections, Torsvik et al., 2012) or methods jointly evaluating the characteristics of multiple 150 constraints including plate velocities, hotspot chains and trench migration (Tetley et al., 2019). 151 Regardless of which method is chosen, the result is a global reference frame defined as a sequence of 152 absolute motions of the African plate, which, together with the relative motions between plate-pairs 153 arranged within a hierarchy (Fig. 2), define the absolute motions of all plates. An exception to all plates 154 being tied to Africa occurs for the Mesozoic . Before 83 Ma (and the opening of the West 155 spreading centre) the motion of the is preserved and reconstructed absolutely to the 156 spin axis through hotspot motion, rather than through Africa. Plate models can also be described in a purely 157 relative framework, in which case a single (or plate) is fixed to its present-day position and all 158 other continents or plates are rotated relative to the fixed plate. This approach is commonly used for 159 localised studies or to easily highlight the difference between two contrasting models (Fig. 2c). 160 161 Before the Mesozoic , it is not possible to use preserved, in situ oceanic lithosphere, hotspot motions 162 and seismic imaging of slabs. Therefore, the logical arrangement of connections within the rotational 163 framework changes (Domeier and Torsvik, 2017), however the general principles of plate reconstructions 164 remain the same. For these times, the only quantitative information on the positions of plates is through 165 palaeomagnetism, which describes the palaeolatitude of continents with respect to the Earth’s spin axis. In 166 these cases, where the absolute motions are more directly constrained than relative motions, rotation models 167 traditionally favour a simpler hierarchy, rather than the complex hierarchies used for post-Pangea times (e.g 168 Fig. 3a, b). The motion of major continents relative to the spin axis is determined using their own 169 palaeomagnetic data (Fig. 3a, b). Within this data set, continents can be grouped together in localised 170 hierarchies where there is evidence that they have remained together or close to one another or, in instances 171 where paleomagnetic data are lacking from some of the blocks if geological constraints permit. 172 173 Finally, although Euler’s rotation theorem is based on the motion of rigid bodies, this is a simplification 174 because the lithosphere is deformable. Recent advancements in plate modelling (Gurnis et al., 2018) have 175 allowed for the development of deforming plate models where rigid plates are able to deform along their 176 edges. 177 178

Figure 2 Overview of the rotational framework of a relative plate motion model. (a) Schematic of the ‘tree’ like hierarchy (e.g. Ross and Scotese, 1988) from the Matthews et al. (2016) reconstruction, where the motion of all plates are defined relative to the African plate at 50 Ma. (b) A simplified map view of the hierarchy in (a). The African stage pole (blue hexagon) is at the top of the hierarchy and connected to the absolute plate motion model that describes the movement of the African plate (in this case, as a proxy for the entire globe) to the deep earth. (c) Example of plate motion with a continent (Africa) fixed in its present- day position. Fixed plate presentations enable observations of relative motions between the fixed continent or plate and another. Hence, they are useful to constrain relative plate motions and also compare two or more different plate models. In this instance, a key difference between the model of Young et al. (2019) and Matthews et al. (2016) is highlighted in the spatial relationship between Laurussia and Gondwana at 410 Ma. In Domeier and Torsvik (2014) (which is preserved in Matthews et al. (2016)) Laurussia is proximal to southwest Gondwana while in Young et al. (2019) it is positioned to the northwest. ADR, Adria; AFR, Africa; ARB, Arabia; AUS, Australia; BAL, (cratonic ); BM, Burma; CAR, ; E.QT, East Qiangtang; EUR, Europe; GRN, ; IC, Indochina; IND, ; JAP, Japan; KOH, Kohistan arc; LHA, ; LHR, ; MAD, ; MBL, Marie Byrd ; MEX, Mexico; NAM, ; NC, North China; NEA, Northeast Africa; NWA, Northwest Africa; NZ, ; PAC, Pacific; PHL, Philippines; PNG, Papua New Guinea; SAL, Salado Microplate; SAM, ; SC, South China; , Southeast ; SIB, ; SOM, Somalia; SPN, Spain; TAU, Taurides; W.QT, West Qiangtang. 179 180 181 2.2 Palaeolatitude 182 183 Palaeolatitude is determined through the study of palaeomagnetic data and is the only method to 184 quantitatively constrain the absolute latitudinal position of a continent for pre- times. If sufficient 185 data are present from a single continent, an apparent polar wander path (APWP) can be constructed that 186 describes the motion of that continent through time (Torsvik et al., 2012). If there are very good geological 187 constraints on the relative positioning between multiple continents, then poles from these continents can be 188 merged to form composite apparent polar wander paths. This merging is done by rotating the poles of 189 multiple continents into the coordinate frame of a single continent, typically found higher in the framework 190 (e.g. Africa, Fig. 2a). If this process is done using all global data it is known as a global apparent polar 191 wander path (GAPWaP). GAPWaPs provide the potential for a more rigorous description of the evolution 192 of a suite of continents as more data are available. 193 194 There are, however, many caveats and uncertainties associated with GAPWaPs and APWPs that contain 195 data sourced from more than one continent (e.g. APWP for Gondwana). In particular, they are strongly 196 dependent on the relative position of continents, as even minor changes in these relative positions can result 197 in large differences in the resulting wander paths. They also are directly dependent on the quality and 198 abundance of palaeomagnetic data, thus a degree of subjectivity can be introduced by what criteria are used 199 for selecting and filtering poles (Van der Voo, 1990; see a new approach in Wu et al., 2020). For example, 200 in the Gondwana APWP of Torsvik et al. (2012) (and also for the APWP we construct in this paper), there 201 is one pole for Gondwana between 440 and 400 Ma that constrains the motion of over half of all known 202 at the time. In the Precambrian, the geological uncertainty of exactly how two (or 203 continents) fit together limits the usefulness of GAPWaPs and APWPs that are defined by multiple 204 continents in conjunction with one another. Instead individual APWPs can be constructed using available 205 palaeomagnetic data for continents and then these are all balanced together globally to produce a coherent 206 kinematic continental model (e.g. Li et al., 2008; Pisarevsky et al., 2014). In this manner, it is possible to 207 build continental drift models purely from palaeomagnetic data. 208

Figure 3 Overview of the rotational framework of an absolute plate motion model. (a) Schematic of a flat hierarchy where the motion of all plates is defined relative to Africa. (b) Map view of the schematic at 450 Ma from Domeier (2018, 2016). Many different plates (blue hexagons) sit at the top of the hierarchy and connect to the deep earth. AFR, Africa; ANT, ; AUS, Australia; BAL, Baltica; C, Carolinia; CY, ; EA, East ; G, ; IC, Indochina; IND, India; K-T, Kazakh-Tianshan; LHA, Lhasa; MAD, Madagascar; NAM, North America; NC, North China; NEA, Northeast Africa; NWA, Northwest Africa; SAM, South America; SC, South China; SIB, Siberia; SOM, Somalia; T, Tarim; WA, West Avalonia. 209 210 2.3 True Polar Wander 211 212 True polar wander (TPW) is the motion of the entire solid Earth (mantle-lithosphere) with respect to the 213 spin axis due to centrifugal forces from Earth’s orbit acting on mass anomalies in the , wherein 214 positive anomalies are driven towards the equator and negative anomalies towards the poles (Evans, 2003). 215 Since TPW is inherent in palaeomagnetic data, all APWP are a composite of both plate motions and some 216 component of TPW. Thus, in the strictest sense, to properly use a plate model for geodynamic modelling, 217 a correction that removes any component of detected TPW should be applied (i.e. the mantle reference 218 frame). Further, as the mass anomalies in the mantle are thought to arise from the flow induced by subducted 219 oceanic lithosphere and the associated return flows, TPW excursions are closely linked to the 220 supercontinent cycle (Zhong et al., 2007). 221 222 Raub et al. (2007) identified three types of TPW summarised briefly below, though only the latter two are 223 relevant here. Type 0 TPW operates on short (< 103 a) timescales as a response to elastic deformation within 224 the lithosphere arising from seismic events (Soldati et al., 2001) and so has a negligible effect on plate 225 motions on the timescales pertinent here (Evans, 2003; Raub et al., 2007). Type 1 TPW is the most 226 important to consider, and is broadly defined as the slow motion of the solid Earth (mantle and lithosphere) 227 readjusting to mass anomalies in the mantle. Type 1 TPW is what is commonly detected and corrected in 228 plate motion models (Steinberger and Torsvik, 2008; Torsvik et al., 2012) in order to constrain geodynamic 229 relationships between deep Earth processes and tectonics (Mitchell et al., 2012). Type 2 TPW (Inertial- 230 Interchange True Polar Wander, IITPW) was originally described by Kirschvink (1997) and is a hypothesis 231 where the mantle and crust are rapidly displaced over large distances relative to the spin axis as the moment

232 of minimum inertia (Imin) approaches the maximum moment of inertia (Imax) resulting in an interchange

233 between the two (i.e. Imin becomes Imax and vice versa). IITPW has been linked to supercontinent breakup 234 as continental lithosphere and subduction zones move away from the , known as a superswell, 235 developing beneaththe supercontinent, which is assumed to remain quasi-stable (Li et al., 2004). In the Li

236 et al. (2004) model, the superswell is likely to maintain the Imax, but as all the continents move away Imin

237 approaches Imax, thus they are speculated to interchange with each other. 238 239 The primary challenge of subtracting the effects of TPW from APW paths is identifying and separating 240 TPW components from continental motion. Methods of detecting TPW vary, either isolating it directly 241 from palaeomagnetic data (Mitchell et al., 2012) or deconstructing the APWP of continents by comparing 242 the kinematic motions of all continental lithosphere on the globe to isolate TPW (Steinberger and Torsvik, 243 2008). Both approaches require a priori assumptions stemming from the choice of plate model being 244 analysed, which makes it impossible to apply a TPW correction from one model to another. For example, 245 fitting a great circle to palaeomagnetic data following the approach of Mitchell et al. (2012) is dependent 246 on knowing the relative continental configuration from which the palaeomagnetic data are sampled. 247 Alternatively, deconstructing the motions of the continents after Steinberger and Torsvik (2008) is 248 dependent on constraining absolute palaeolongitude to separate TPW from apparent polar wander. In order 249 to support their arguments, both methods are dependent on having some form of absolute palaeolongitudinal 250 control. 251 252 2.4 Palaeolongitude 253 254 There is no well-established method to compute the absolute palaeolongitude of any given plate that is 255 applicable to pre-Jurassic times. However, two hypotheses have been proposed for establishing absolute 256 palaeolongitude for pre-Pangea reconstructions (Torsvik and Cocks, 2017): (i) the plume generation zone 257 method (PGZ) (Torsvik et al., 2014) and (ii) the orthoversion model of the supercontinent cycle (Mitchell 258 et al., 2012). 259 260 The PGZ method is based on reconstructing the surface locations of and large igneous provinces 261 (LIPs) to the margins of the large low shear velocity provinces (LLSVPs) (Burke and Torsvik, 2004; 262 Torsvik et al., 2010a) situated on the core-mantle boundary (CMB) (Garnero et al., 2007; Li and McNamara, 263 2013). The foundation for this method is the observation that kimberlites and LIPs, when restored to a 264 mantle reference frame at the time of eruption, are positioned preferentially above the margins of the 265 LLSVPs (Burke and Torsvik, 2004; Torsvik et al., 2010a). LLSVPs are regions of anomalously slow 266 (slower than ambient mantle) seismic velocities, close to the core–mantle boundary. Since large-scale 267 mantle structure is intimately related to the supercontinent cycle, LLSVPs (Li and Zhong, 2009) are thought 268 to exist for (at least) a similar time frame as the supercontinent itself (i.e. present-day LLSVPs are thought 269 to have existed at least back to 320 Ma) (Li and Zhong, 2009), but not necessarily maintaining their present- 270 day geometry (Flament et al., 2017; Zhang et al., 2010; Zhong and Rudolph, 2015). 271 272 The PGZ method to establish absolute palaeolongitude therefore assumes: (i) the geometric stability of the 273 present-day positions of LLSVPs back to the time of interest, and (ii) that there is a positive statistical 274 correlation with margins of LLSVPs and the extrusion of LIPs and kimberlites at the Earth’s surface. 275 Adopting the PGZ method allows one to position continents longitudinally by reconstructing the positions 276 of LIPs and kimberlites to overlie the margins of LLSVPs while simultaneously utilising TPW-corrected 277 palaeomagnetic data (Torsvik et al., 2014). In this way, models with explicitly defined and reproducible 278 absolute plate motions can be created (Domeier and Torsvik, 2014). However, a number of recent studies 279 have raised questions about the assumptions implicit in the PGZ approach. For example Flament et al. 280 (2017) and correspondence by Torsvik and Domeier (2017), Doucet et al. (2020), Zhong and Rudolph 281 (2015) and Zhong and Liu (2016) on long term LLSVP stability and Austermann et al. (2014), Davies et 282 al. (2015), with response by Doubrovine et al. (2016) on the statistical correlation. Consequently, while 283 utilising the PGZ method for plate reconstructions in the Palaeozoic is currently an open area of research, 284 the theoretical and practical application for Precambrian times remains untested. 285 286 The other proposed method of determining absolute palaeolongitude for deep time reconstructions is known 287 as the ‘orthoversion model’ and suggests that successive supercontinents coalesce orthogonally (90° 288 longitude) above the downwelling formed by subduction at the margin of the previous supercontinent 289 (Mitchell et al., 2012). Mitchell et al. (2012) test their model by first determining the minimum moment of

290 inertia (Imin) during each phase of supercontinent assembly. This is done by rotating the available 291 paleomagnetic data into a relative reference frame of a fixed continent (Africa for the Phanerozoic,

292 for the Neoproterozoic) and then fitting a great circle to the resulting poles. Imin is defined as the 293 orthogonal axes (or pole) of the great circle and is taken to approximate the TPW axis. Mitchell et al. (2012) 294 fitted great circles to palaeomagnetic data for a selection of time periods (1165–1015 Ma, 805–790 Ma,

295 550–490 Ma and 220–90 Ma) and calculated that the angle between each successive Imin was ~90°; as 296 expected in the orthoversion model. The initial (palaeolongitudinal) placement of Pangea is constrained

297 from its position at ~90 Ma (centroid of Imin at 0°N, 10°E) when the most recent TPW episode finished

298 (Steinberger and Torsvik, 2008). Therefore, during Rodinia and Gondwana, when Imin was 90° from present 299 day, Mitchell et al. (2012) proposed that the centre of mass of both was positioned at 100°E. 300 301 As with the PGZ, a number of limitations are apparent with the orthoversion model. In particular, the 302 method of Mitchell et al. (2012) does not separate continental motion from true polar wander when

303 calculating Imin, thus it assumes that during these times TPW is the primary signal recovered from 304 palaeomagnetic data and not continental motion. Secondly, the orthoversion model is inherently dependent

305 on both the quality and abundance of palaeomagnetic data (e.g. the Rodinia Imin in Mitchell et al. (2012) is 306 based on only three poles), as well as the continental configuration of the time (as the continental 307 configurations can determine the relative position of palaeomagnetic data when rotated into a specific 308 reference frame). Torsvik and Cocks (2017) highlight this succinctly by using slightly different

309 palaeomagnetic data, and a slightly different Gondwana configuration to produce an Imin between 550 and 310 490 Ma of 50°S, 64°E (compared to the estimate of Mitchell et al. (2012) of 30°S, 75°E). Finally, the

311 calculation of Imin at each time step occurs within a fixed relative reference frame (i.e. Fig. 2c), meaning

312 that the Imin itself cannot be restored to an absolute palaeogeographic position. Thus, the longitudinal centre 313 of Rodinia at 100°E is not explicitly proven by Mitchell et al. (2012). Instead, since the calculated

314 successive Imin’s are 90° apart, it is inferred to be in this location. 315 316 2.5 Geology 317 318 Geology is unfortunately silent on the absolute positioning of continents in time, except in circumstances 319 such as the PGZ method discussed above. However, it contains a wealth of information of relative plate 320 motions in the Mesozoic and Cenozoic (e.g. ), and in deeper time through the temporal 321 evolution of sedimentary basins and facies, and tectonic affinities inferred from , 322 arrays, magmatism and metamorphism. However, geology does have an advantage over palaeomagnetic 323 data in that there are many different types of data available, especially from small and minor terranes in the 324 Precambrian that are otherwise unconstrained palaeomagnetically. Given the wealth of geological data, 325 especially in the Neoproterozoic, there is a general hierarchy of use that is a reflection of the scale of the 326 problem. Our approach is to start by building a global framework and work progressively to finer resolution 327 to inform the localised nature of that framework. 328 329 The first and most important geological data to gather is evidence of and arcs, as they describe 330 separation (usually leading to seafloor spreading) and convergence, respectively and can therefore put in 331 place the framework for plate motions and subsequent interpretation of geological data. Secondly, 332 identifying piercing points where geological boundaries can be matched on now separate continents (e.g. 333 Appalachians and Gondwana forming ), or ways of fitting two continents together in a stable 334 configuration. These piercing points are important because reconstructing continental configurations with 335 high confidence allows the generation of more rigorous APWPs and they help to constrain the location and 336 orientation of both rifts and arcs on the periphery of continents. Unfortunately, due to deformation and 337 progressive alteration of continental crust (e.g. changes of continent-ocean boundaries (COBs)), in pre- 338 Gondwana times it is difficult to be more precise than matching the margins of large continents and for 339 smaller cratons and terranes it is almost impossible. A pertinent example of this are the four different 340 proposed configurations of Australia and Laurentia during Rodinia (see reviews by Li et al., 2008; Merdith 341 et al., 2017b), which all broadly match the same margins against each other (east coast of Australia with 342 the west coast Laurentia, with or without an intervening continent) but place them in different relative 343 positions. 344 345 After arcs and rifts, we can loosely (but not exhaustively) group geological features and data based on their 346 applications. Sedimentary basins, dyke swarms, detrital minerals, geochemical signatures and are 347 typically used to determine or latitudinal band and align once contiguous regions. The time- 348 scales and conditions of metamorphism together with structural data can be used to infer the tectonic setting 349 and polarity of collisional events or help constrain the nature of indeterminate plate margins such as 350 transform boundaries. These types of data assist in increasing the resolution of a plate model by 351 understanding the geology at smaller scales within the framework of a specific tectonic setting, such as an 352 arc or . We stress here that the relationship between detailed and the broader framework 353 of a plate model is not a ‘one-way street’ but is highly iterative. If, for example, a detailed geological study 354 determines that an interval of magmatism and sedimentation that was originally interpreted as a failed rift, 355 in fact led to seafloor spreading, then the broader scale tectonic framework and plate model must be re- 356 evaluated. Finally, due to the qualitative nature of most geological data, the iteration and implementation 357 of these data into the plate model typically necessitates qualitative decisions that others may disagree with. 358 Iteration over the model continues until we approach tectonic congruency within the model (i.e. data-based 359 iterations in one part of the world do not nullify data in other areas of the world). We stress that this does 360 not mean our model is ‘correct’ or ‘true’, just that it is internally consistent with as much data as possible. 361 Consequently, we consider the model presented here a viable, but non-unique interpretation of 362 Neoproterozoic data. 363 364 A specific example of the importance of interpreting geology within a full-plate framework is given by 365 recent work on the – evolution of the East African Orogen. Although arc-related 366 magmatism has been recognised in the northern East African Orogen for a number of decades (e.g. Stern, 367 1994), the recognition of similar-aged arc magmatism in the higher-grade southern East African Orogen of 368 Madagascar, Southern India and has been more controversial and under-appreciated until 369 recently (Archibald et al., 2018, 2017; Armistead et al., 2019; Plavsa et al., 2015; Ruppel et al., 2018). In 370 one particular example, work over the last decade on Western Dronning Maud Land in East Antarctica has 371 identified an extensive Stenian– juvenile arc system (named TOAST; (Elburg et al., 2015; Jacobs et 372 al., 2015; Ruppel et al., 2018)). This discovery has gone hand-in-hand with the recognition of a similar 373 in western Madagascar, known as the Dabolava Arc (Archibald et al., 2017; Tucker et al., 2011) 374 (e.g. Fig. 8). These arcs are now separated by a considerable distance, but their reconstructed position in 375 the Neoproterozoic and their similarity has led to us interpreting them as part of one continuous subduction 376 system that was active for the Stenian and Tonian. In this manner, we now include TOAST as another part 377 of Azania and have reworked a number of the plate boundaries in Merdith et al. (2017a) to reflect these 378 new geological data and tectono-geographic interpretations. 379 380 2.6 Kinematic considerations 381 382 The final line of reasoning used to create full plate models are plate kinematic constraints. These are not 383 defined explicitly through geological or geophysical data of the types outlined above, but rather come from 384 the idea that the evolution of plate motions through time must follow the broad principles of plate tectonics 385 in a way that would seem reasonable; for example, by equivalency with more recent and well-constrained 386 plate motions. The most basic requirement is that continental blocks cannot pass through or significantly 387 overlap other continents and we must be able to describe the position and motion of each continental block 388 for as long as the crust within that block is thought to have existed. While this may seem obvious, these 389 considerations present a powerful method for discriminating between competing reconstruction scenarios. 390 Models constructed for deep time that cannot evolve towards more recent and present-day configurations 391 of continents cannot be considered correct. Similarly, models requiring an implausible kinematic evolution 392 in order to meet present-day configurations cannot be correct. A tangible example is Rodinia, for which a 393 range of configurations could be permissible based on available paleomagnetic and geological data (see 394 reviews in Evans, 2013; Li et al., 2008; Merdith et al., 2017b). However, analysing the sequence of plate 395 motions required to translate each continent to their (better constrained) positions during the Palaeozoic is 396 more plausible in some of these scenarios than others, such as not requiring individual terranes or blocks to 397 cross multiple ocean basins or navigate their way around a stable continent (Merdith et al., 2017b). Further 398 examples where kinematic constraints add useful insights are when constructing models that explicitly trace 399 the evolution of plate boundaries and tectonic plates. An example is expressed in the logic of Domeier 400 (2018) who inferred that the longitudinal position of Tarim, North China and South China during the late 401 Cambrian– must have remained stable relative to one another, because palaeolatitudes from 402 palaeomagnetic data overlapped and therefore must be consistent with their end position in the more well 403 constrained Devonian–. 404 405 Finally, we take a uniformitarian view of tectonic evolution, in that we assume that plate tectonics and 406 relative plate motions were operating on similar principles in the Neoproterozoic to what we can observe 407 in the Mesozoic and Cenozoic. During more recent times, the motion of plates remains relatively constant 408 for time lengths on the order of 10–100 Ma, with changes in motion occurring comparatively quickly (< 3 409 Ma) and tied to an event further afield, such as collision in a subduction zone, subduction onset, 410 rifting onset or ridge subduction (e.g. Austermann et al., 2011; Cawood et al., 2016; Cawood and Buchan, 411 2007; Knesel et al., 2008). The representation of this within a conceptual framework is that a single 412 continent or plate may move for time lengths on the order of 10–100 Ma around a single Euler pole, before 413 a plate re-organisation event triggers a change in direction and velocity of the plate (Gordon et al., 1984). 414 Within palaeomagnetic data, this approach is exemplified in Torsvik et al. (2008), where filtered 415 compilations of data result in smooth APWP segments punctuated by cusps in motion and velocity, and 416 more recently by Wu et al. (2020). For the Neoproterozoic, which has much more sparse palaeomagnetic 417 data coverage than the Phanerozoic, the logic can be applied by linking changes in plate direction (as 418 suggested or necessitated by palaeomagnetic data) directly to geological evidence of a change in tectonic 419 regime within the region of interest (e.g. Merdith et al., 2017b). 420 421 2.7 Synthetic Ocean Plates 422 423 A complete full-plate model by definition includes a representation of the evolution of ocean basins through 424 time. This is the most uncertain part of any full-plate model. With a few exceptions (e.g. Granot, 2016), no 425 pre-Jurassic ocean crust is preserved in situ, so that even where we can infer the presence of divergent plate 426 boundaries (for example, following continental breakup), the precise geometry and spreading rates at these 427 boundaries are conjectural. Consequently, there are no unique solutions to the definition of these boundaries 428 and the synthetic ocean plates constructed from them. The main aim in the reconstruction of ancient ocean 429 basins is to ensure that the synthetic plates and plate boundaries are at least consistent with sparse 430 observations preserved on the continents. 431 432 Firstly, although we have little data on the creation of oceanic lithosphere for pre-Jurassic times, the 433 geological record does preserve data on the consumption of oceanic lithosphere at subduction zones. The 434 most important criteria that therefore must be met are that (i) (sub-) orthogonal divergence occurs at mid- 435 ocean ridges and that (ii) convergence occurs at subduction zones (Domeier and Torsvik, 2014). The former 436 of these two criteria typically occurs between continents during supercontinent breakup or when small 437 terranes rift from a continent (Dalziel, 1997). The latter of these includes subduction along the margins of 438 continents as well as within intra-oceanic arcs. Thus, these two criteria necessitate the extrapolation of 439 known plate boundary positions (e.g. preserved continental arcs, rift zones) into larger areas to ensure the 440 tectonic congruency of the model (in this case, that convergence or divergence at one location doesn’t 441 nullify the same criteria in another location). It is this step, in particular, that requires significant iteration 442 when constructing a model. Maintaining tectonic congruency for a model is best achieved when the 443 extrapolation of plate boundaries is done as simply as possible. For example, in reconstructing an ocean 444 basin without any continental crust (e.g. Pacific Ocean, Ocean), a is usually the 445 simplest expression of a ridge system that ensures divergence in all directions and convergence at its 446 margins (Domeier and Torsvik, 2014). The evolution of such a triple junction could also be seen in the 447 opening of the (e.g. Pisarevsky et al., 2008; Robert et al., 2020), 448 449 3 Model selection and justification 450 451 3.1 Existing plate models 452 453 The selection of which full-plate model to assist in solving a problem is dependent on the nature of the 454 problem, as each published plate model is constructed using a different approach and has a different 455 reconstruction framework. For example, a study looking at absolute plate motions of the Cenozoic has little 456 use for models connected to the Palaeozoic or Neoproterozoic. Instead, such studies typically include a 457 comparison with previously published models as well as a rigorous mantle reference frame. Similarly, a 458 study that traces the latitudinal distribution of continental arcs through the Phanerozoic has no need for a 459 mantle reference frame and a study investigating the changes in net rotation through time would be 460 concerned with small, localised improvements from regional models but more focussed on capturing the 461 large scale changes that occur when continents breakup. Although newer plate models typically address the 462 shortcomings of previous models or implement more refined updates of regional areas, this does not 463 necessarily make them better for all applications. Older plate models have been more rigorously tested and 464 used by the community and as one travels further back in geological time, the data are more ambiguous and 465 can invite alternative interpretations. For the purpose of this study, the existing plate models we consider 466 are only those that are publicly released with fully self-consistent with coherent plate motions and plate 467 boundaries. Thus, we omit many models that provide only continental motions (Scotese, 2016; Torsvik and 468 Cocks, 2016), alternative or regional Rodinia configurations (Evans, 2013) or regional refinements of 469 global models for the Mesozoic and Cenozoic (e.g. Vaes et al., 2019). 470 471 The major step forward for producing full-plate models was the construction of open-source computer 472 software specifically designed to work with full-plate reconstructions (e.g. GPlates; (Gurnis et al., 2012). 473 Following their development, GPlates-compatible global models for the to present (Seton et 474 al., 2012; Shephard et al., 2013) and a model for the Late Palaeozoic (Domeier and Torsvik, 2014) soon 475 followed (hereafter, SET12 and DT14, respectively). Subsequent work by Matthews et al. (2016) (MAT16) 476 bridged the gap between the Palaeozoic and Jurassic, linking a slightly modified version of DT14 with an 477 updated SET12 model, the Müller et al. (2016) model (MUL16). Further back in time, two models for the 478 Early Palaeozoic (500 to 410 Ma) now exist: Domeier (2016) (DOM16), which encompasses the evolution 479 of the Iapetus and Rheic Oceans, as well as the motion of Gondwana, and Domeier (2018) (DOM18), which 480 models the evolution of the first generations of Tethyan Oceans and Central Asian blocks (Siberia, North 481 and South China, Tarim). For the Neoproterozoic, Merdith et al. (2017a) (MER17) produced a full-plate 482 model from 1000 to 520 Ma, using the models of continental motion presented by Li et al. (2013, 2008) as 483 a base. An alternative reconstruction from the late Palaeozoic to present-day has been presented by Young 484 et al. (2019) (YOU19). YOU19 offers an alternative full-plate model for the Palaeozoic to the DT14 and 485 MAT16 models that does not rely on the PGZ method. Finally, a deforming plate model was produced 486 (Müller et al., 2019) (MUL19) that modelled rift and convergence deformation from 250 to 0 Ma. Table 1 487 summaries the main features of these models. 488 489 490

Figure 4 ‘Family-tree’ of full-plate models and some of the key underlying continental drift type models leading to their development. See text for more details. The model presented here bridges the MER17, DOM16, DOM18 and YOU19 models into a single continuous plate model. X-axis not to scale. Cr, Cryogenian, and; Ed, Ediacaran. 491 492 3.2 Cenozoic and Mesozoic plate models 493 494 We consider three plate models for the Cenozoic and Mesozoic: SET12, MUL16, and MUL19 (Fig. 4). The 495 SET12 model spans from present-day to 200 Ma, with MUL16 and MUL19 extending back to 230 and 240 496 Ma, respectively. The main geological constraint of these models are the magnetic lineations preserved in 497 that describe the relative movement between the pairs of continents breaking apart during the 498 fragmentation of Pangea. The evolution of the Atlantic, Indian, Southern, and Cenozoic Pacific 499 oceans are consistent, to the first order, across all three models. Larger differences between the models arise 500 in regions where oceanic crust has been subducted, upon which they then rely on a combination of 501 geometric (e.g. assumption of symmetrical spreading), geological, seismic and palaeomagnetic data to 502 constrain the motion and evolution of terranes that open and close ocean basins (e.g. Liu et al., 2010, 2008; 503 Sigloch and Mihalynuk, 2013). In these regions, the tectonic histories, even for Cenozoic times, remain the 504 subject of ongoing research and the scenarios embedded in the global models used in this study represent 505 one candidate amongst many competing models. For example, the extent of subduction systems within the 506 Tethyan domain and the nature of India- collision is still hotly debated (Hu et al., 2016; Parsons et 507 al., 2020; van Hinsbergen et al., 2020, 2012). Similar combinations of methods have been used to propose 508 alternative interpretations for circum-Pacific regions, especially for the and earlier times in the 509 northwest Pacific (Domeier et al., 2017; Konstantinovskaya, 2002; Shapiro and Solov’ev, 2009; Vaes et 510 al., 2019; Wu et al., 2016), the southwest Pacific (Hochmuth et al., 2015; Matthews et al., 2015; Schellart 511 et al., 2006; Sutherland et al., 2020) and the northeast Pacific (Clennett et al., 2020; Sigloch and Mihalynuk, 512 2013). All of these regions invite competing models and alterations to existing global models and it is 513 notable that many of the studies mentioned above have benefited from using the resources made available 514 by previous global studies, beginning with SET12, as a starting point for detailed regional analysis. 515 516 All three models employ a relative hierarchy (Fig. 2a), in which a fully relative plate motion model is tied 517 to an absolute plate motion model through Africa. The relative hierarchies are similar across the models 518 because of the preserved oceanic crust. These global relative motion hierarchies are then linked to an 519 absolute reference frame tied to the mantle, though the details of these reference frames differ between 520 models. SET12 uses a hybrid reference frame, using a moving hotspot reference frame for 100–0 Ma 521 (O’Neill et al., 2005) and a true-polar wander corrected palaeomagnetic reference frame for 200–100 Ma 522 (Steinberger and Torsvik, 2008). MUL16 also adopts a hybrid absolute reference frame, but uses the 523 moving hotspot model of Torsvik et al. (2008) instead of that of O’Neill et al. (2005) based on an assessment 524 of the geodynamic plausibility of a range of alternative mantle reference frames by Williams et al. (2015). 525 MUL19 departs from both SET12 and MUL16 in that is uses an absolute reference frame derived by Tetley 526 et al. (2019). The method of Tetley et al. (2019) optimises absolute plate motions through a joint inversion 527 involving trench motion, fitting hotspot motion tracks and net rotation to determine the motion of Africa 528 (at the top of the relative hierarchy) that simultaneously best fits all three criteria. Despite the emphasis on 529 mantle reference frames in these previous global models, the same relative plate motion hierarchies can 530 also be tied to a pure paleomagnetic reference frame (e.g. Cao et al., 2019; Torsvik et al., 2008). 531 532 In this study, we rely on palaeomagnetic data as the main basis for linking absolute plate configurations 533 continuously from the Cenozoic to the early Neoproterozoic, when tying plate configurations to the mantle 534 is far more problematic. Nonetheless, some aspects of the more recent plate motions still rely on 535 observations from hotspot trails—specifically, the motion of the Pacific Plate and other oceanic plates that 536 have bordered it. During the , these plates lay within the Panthalassa ocean basin that was 537 entirely surrounded by subduction zones, meaning that we cannot tie the motions of the oldest crust of the 538 Pacific Plate to the continents by seafloor spreading anomalies. 539 540 Finally, with regard to Mesozoic-Cenozoic global plate models, we note that MUL19, while containing the 541 same relative framework as SET12 and MUL16, also contains deforming plates. In MUL19, deformation 542 of rifts and collisional zones are modelled explicitly, making it the first plate model to step away from the 543 simplification of rigid plates that all other models assume. However, the reconstruction here relies on the 544 simpler, rigid approximation used in SET12, MUL16, and other previous studies. 545 546 3.3 Mid-late Palaeozoic plate models 547 548 The progression of the three Palaeozoic plate models (all modelling 410–250 Ma) is slightly more nuanced 549 than in the Mesozoic and Cenozoic because of the absence of preserved oceanic crust. DT14 is the original 550 model and both MAT16 and YOU19 used DT14 as the basis of their Palaeozoic models and then connect 551 to MUL16 for the Mesozoic and Cenozoic. In effect, MAT16 is the connection of DT14 (i.e. minimal 552 changes) to MUL16, while YOU19 is an alternate version of DT14 for the Palaeozoic. 553 554 DT14 is heavily based on work by Torsvik et al. (2012) and models the amalgamation of Pangea through 555 the collision of Laurussia and Gondwana and the evolution of the Palaeo-Tethys and opening of the Meso- 556 Tethys oceans. The model has a flat hierarchy (Fig. 3a) with APWPs defined for each individual continent 557 and each continent being tied directly to the spin axis. Domeier and Torsvik (2014) also use the PGZ method 558 to constrain absolute palaeolongitude. Therefore, their model assumes the stable, immutable nature of the 559 present-day LLSVPs within the mantle back to 410 Ma. DT14 is presented in both a mantle and a 560 palaeomagnetic reference frame, with the mantle reference frame being corrected for TPW after Torsvik et 561 al. (2014) . 562 563 MAT16 adopted the DT14 model, with minor amendments required to link it with MUL16 (see Matthews 564 et al. (2016) for details). A key difference between the two models is that MAT16 translated the flat 565 hierarchy of DT14 into a fully relative reference frame (i.e. converted the structure from Fig. 3a into Fig. 566 2a), where the motion of all plates is tied to Africa, which is then tied to an absolute plate motion model. 567 The absolute plate motion model of MAT16 is the same as in both MUL16 (i.e. a hybrid between hotspots, 568 slabs and palaeomagnetic data) and DT14 (i.e. absolute latitude and longitude, corrected for TPW). Thus, 569 MAT16 also invokes the PGZ method. 570 571 YOU19, while starting from MAT16 as a base, differs much more from MAT16 than MAT16 does from 572 DT14. This is because YOU19 uses a different base assumption, leading to notable changes in the actual 573 plate model itself. The most important difference is that YOU19 does not assume that LLSVPs were fixed 574 and stable back to 410 Ma. They abandon the PGZ method for constraining palaeolongitude and thus argue 575 that they can better accommodate geological and kinematic (plate speed and trench migration) criteria more 576 strongly than either DT14 or MAT16. The two key changes that YOU19 implemented (relative to DT14 577 and MAT16) are shifting Laurussia in latitude and longitude to be closer to its position in Pangea against 578 Gondwana, thus removing a dextral motion between the Patagonian margin of South America and Laurussia 579 (e.g. Fig. 2c), and removing easterly drift of South China during the –. The 580 implementation of both in DT14 (and then in MAT16) is a consequence of the PGZ method, since the 581 longitudinal position of Laurussia and South China is based on fitting preserved eruptions to the edges of 582 LLSVPs (Domeier and Torsvik, 2014). YOU19 argue that the interpretations of DT14 introduce unrealistic 583 kinematic scenarios: 8000 km of relative dextral motion between Laurussia and the Patagonian margin of 584 Gondwana (e.g. Fig 2c) at a relative plate velocity of 30 cm/a and South China moving at plate speeds of 585 40 cm/a between 260 and 250 Ma in MAT16 (Young et al., 2019). The dextral motion between Gondwana 586 and Laurussia of DT14 and MAT16 that that YOU19 removed is not a transition from the Pangea B to 587 Pangea A configuration, which is explicitly defined as dextral motion after Pangea formed (Domeier et al., 588 2012). Rather, all three models adopt a Pangea A configuration and include some component of dextral 589 motion between Laurussia and Columbian–Mexican margin of Gondwana in the Devonian. However, the 590 position of Laurussia in YOU19 is unsupported palaeomagnetically by ~30° (Section 4.2 and 5.1), which 591 is problematic as there is an abundance of data from both Laurentia and Baltica to constrain its position at 592 this time. 593 594 Recently, Wu et al. (2020) have also proposed an integrated geological and palaeomagnetic model for the 595 amalgamation of Pangea. Their study used a different selection of palaeomagnetic data (all three mid-late 596 Palaeozoic full-plate models discussed here used the compilation of Torsvik et al. (2012) as a base) and a 597 new method of APWP generation that weighted poles based on their quality and uncertainty. The model of 598 Wu et al. (2020) also suggested that the formation of Pangea was originally initiated by collision between 599 Laurussia and a promontory of Gondwana consisting of the Variscan Massifs at ca. 400 Ma. In their model, 600 the promontory formed by the scissor-like opening of the Palaeotethys Ocean off the northern margin of 601 Gondwana. In principle this model would be very compatible with the one we present here, however as the 602 model of Wu et al. (2020) is currently only palaeogeographic, we do not consider it as an option for merging 603 in this study. 604 605 YOU19, like MAT16, uses a relative plate hierarchy with Africa as the root of the hierarchy. Africa is 606 connected to an absolute plate motion model using the Torsvik and Van der Voo (2002) APWP for the 607 Palaeozoic, before the model transition to the absolute plate motion model of MUL16. Because the model 608 has no absolute palaeolongitude control, YOU19 does not constrain or correct for TPW in the Palaeozoic 609 and is therefore presented in a purely palaeomagnetic reference frame. 610 611 3.4 Early Palaeozoic plate models 612 613 Two separate models exist for part of the Early Palaeozoic between 500 and 410 Ma. Each of these two 614 models focus on a separate area of the Earth at the time. DOM16 focuses on the evolution of the Iapetus 615 and Rheic oceans and the amalgamation of Laurussia (Baltica, Laurentia and Avalonia). DOM18 models 616 the evolution of Siberia, Gondwana, the terranes that now make up the Central Asian and 617 the Chinese cratons (Tarim, North and South China). While each model focuses on a different area, the 618 overarching assumptions and framework of both models are identical. Both models follow the approach of 619 DT14, possessing a flat hierarchy with APWPs being defined for most continents, such that they all move 620 independently from each other. The model extends the assumptions of the PGZ method back to 500 Ma, 621 using the location of LIPs and kimberlites to constrain absolute palaeolongitude. Both models therefore 622 have a TPW correction, and are presented in a mantle and palaeomagnetic reference frame. 623 624 3.5 Neoproterozoic 625 626 Only one full plate model exists for the Neoproterozoic (MER17, Merdith et al., 2017a). MER17 is based 627 on Li et al. (2013, 2008) and models the evolution of Rodinia; it’s breakup and the amalgamation of 628 Gondwana (1000–520 Ma). There are, however, a number of important considerations that differentiate 629 MER17 from Phanerozoic full-plate models. Firstly, MER17 used a hybrid plate rotation hierarchy, 630 defining two separate nodes that move independently from each other using palaeomagnetic data tied 631 directly to the spin axis (India as one and Laurentia, as the centre of Rodinia, as the second) (Fig. 5a, b). In 632 this model, India and Laurentia both act as separate roots that then constrain a of relative plate 633 rotations that collectively describe the rest of the world. Secondly, the continental drift model of Li et al.

634 (2013) invokes the orthoversion model of determining palaeolongitude, where they fix the Imin of Rodinia 635 to be at 100°E. As MER17 adopted these rotations as their base, there is an element of the orthoversion 636 model preserved between models. However, MER17 also drastically changed the configuration of Rodinia, 637 as well as the timing of breakup, compared to Li et al. (2013), relative rotations within the Laurentian (i.e. 638 Rodinian) node and absolute rotation of Laurentia itself in order to fit geological and kinematic constraints.

639 They did not recalculate TPW and the Imin of Rodinia, thus MER17 does not have a strict absolute 640 palaeolongitude control. Finally, there is no correction for TPW in MER17 and no mantle reference frame; 641 the model is presented purely in a palaeomagnetic reference frame. 642

Figure 5 Overview of the rotational framework of a hybrid plate motion hierarchy. (a) In this case, two different nodes or clusters sit at the top of their respective branches of the hierarchy and ground regional relative plate motions. (b) Map view of the schematic at 900 Ma from this study. Two different plates (India and Laurentia; blue hexagons) sit at the top of the hierarchy and connect to the deep earth, with regional plate motions being defined relative to each. 643 644 3.6 Model selection 645 646 In order to produce a coherent global model, we must select from the models described above that best 647 align with our goals: (i) open the Neoproterozoic and Cambrian up for quantitative tectonic analysis; (ii) 648 create a framework that can support local or regional studies for the Neoproterozoic and (iii) a foundation 649 for future studies looking at long timescale (> 10–100 Ma) trends in either tectonics or geodynamics. 650 651 As our aim is to produce a model back to 1 Ga, three choices are already made for us because they are the 652 only models that exist for those time periods: MER17 for 1000–520 Ma, and DOM16 and DOM18 for 500– 653 410 Ma. For the remaining time period, MAT16 (an extension of DT14) and YOU19 are viable options. 654 Both models link to the MUL16 model and choosing between them requires considering the reconstruction 655 framework (e.g. hierarchy, reference frame) of our model with respect to our intent. To satisfy our goals, 656 we need a model in a palaeomagnetic reference frame. We acknowledge the value and potential in exploring 657 hypotheses for constraining palaeolongitude. However, since tectonic reconstruction models, such as the 658 one we are presenting here, are a required starting point for exploring the long-term evolution of mantle, 659 we have kept the model conservative by not assuming the fixity of LLSVPs or that TPW dictates 660 supercontinent position. Therefore, we elect to leave palaeolongitude unconstrained by either PGZ or 661 orthoversion methods. Thus, we selected the YOU19 model, as it has removed the absolute palaeolongitude 662 controls adopted in MAT16. We stress that our approach is deliberately conservative and requires the fewest 663 a priori assumptions, but anticipate that future developments will see improvements in the model presented 664 and that a comparison of several end member models would be useful for evaluating the long-term 665 connection between lithosphere and mantle. One promising way forward could be to optimise tectonic 666 parameters, such as subduction zone migration and plate velocity in the manner of Tetley et al. (2019) to 667 define a mantle reference frame. The model we present is an essential precursor for such techniques. Our 668 model is constructed with a palaeomagnetic reference frame, that is, even at younger times, there is no 669 mantle reference frame. For studies needing such a reference frame, which only exists since the Cretaceous, 670 we suggest people use either SET12, MUL16 or MUL19. 671 672 3.6.1 Our approach to plate modelling 673 674 Here we outline our approach to merging individual plate models into a coherent global plate model 675 spanning the past 1 Ga. As detailed above, our goal is a geologically constrained model within a 676 palaeomagnetic framework. In addition to reconstructing plate motions, we also model plate boundaries, 677 which requires us to also focus on the relative motion between plates. The evolution of plate boundaries is 678 commonly preserved in the geological record such as passive margins marking divergent boundaries and 679 magmatic arcs recording convergence. Palaeomagnetic data, although providing a quantitative absolute 680 constraint on the position of a , have varying uncertainty that allow some manipulation and flexibility. 681 For example, continental drift models (and APWPs) typically fit the data as tightly as possible. However 682 recent studies have analysed the effect of exploring both the temporal (i.e. age constraints) and statistical 683 uncertainty to create alternative APWPs (Tetley et al., 2019; Wu et al., 2020). Given this, we use a hybrid 684 approach that adopts parts of both the flat, palaeomagnetic hierarchy traditionally used for Precambrian 685 reconstructions and the relative framework used in more recent times. 686 687 Using this approach, our model has multiple clusters of related cratonic elements moving together (Fig. 5a). 688 Whether a specific cratonic element moves relative to another is dependent on their geological relationship. 689 For example, if they are separated by an incipient ocean basin as indicated by geological data then we 690 suggest that it is easiest for them to be moving relative to each other (within the bounds of whatever 691 available palaeomagnetic data), because the relative relationship of divergence (expressed as a mid-ocean 692 ridge) is preserved. The hierarchy is defined by geological precedence, where terranes move relative to 693 blocks, blocks move relative to cratons and cratons move relative to ‘supercontinents’. In this manner, 694 generally (but not necessarily exclusively), crust with more preserved data should always be placed above 695 crust with less preserved data in the hierarchy, because we have more confidence in the geological evolution 696 (and also likely palaeomagnetic constraints) from these continents. Alternatively, if continents are separated 697 by a large ocean basin, we form a new cluster. This cluster-approach has an added benefit in the more 698 uncertain Neoproterozoic, as it means that we can simply introduce a new cluster for Rodinia, which has 699 Laurentia as the root of the relative hierarchy, instead of maintaining Africa or the at the top 700 of the cluster (which are the roots of Pangea and Gondwana, respectively) and can also introduce a cluster 701 for India and South China which, in our interpretation, move separately to Rodinia on the other side of the 702 globe. 703 704 A full-plate reconstruction models both continents (palaeomagnetic data) and plate boundaries (geological 705 data). Therefore, we use both data simultaneously to iterate towards a solution. Palaeomagnetic data are 706 used initially to build a continental drift framework (e.g. (Li et al., 2013, 2008). If palaeomagnetic data are 707 abundant and of high quality, then either an APWP or GAPWaP are constructed, which we do for the 708 Phanerozoic. We then introduce geological data in the form of plate boundaries (e.g. compilation of rifts 709 and arcs (Merdith et al., 2017b)) and use the compilation to manipulate the model in a manner that still fits 710 the palaeomagnetic data, but also accommodates geological data. Structural and metamorphic constraints 711 are used here principally to infer (where possible): (i) polarity of subduction, (ii) collision timing and (iii) 712 orientation of rifting. Once the broader tectonic framework is implemented, we begin introducing smaller 713 blocks and terranes into this framework. This approach allows us to increase the resolution of the model 714 within key areas while maintaining the tectonic coherency of the model as a whole. For example, where the 715 broader tectonic framework models subduction leading to collision, the introduction of terranes and smaller 716 blocks that preserve evidence suggesting a two-stage collision or of an oceanic arc can be used to 717 more finely model the plate boundary network. Other pieces of geological data, such as faunal provinces 718 (e.g. Burrett et al., 1990), isotopic signatures (e.g. Collins et al., 2011) and detrital (e.g Cawood et 719 al., 1999) are used here to assist with connecting disparate terranes to larger blocks that they share affinity 720 with or to infer the presence of a plate boundary not directly preserved in the geological record (such as by 721 a diverging record). 722 723 Because our model contains geological data in the form of plate boundaries, we are particularly interested 724 in ensuring the forwards and backwards compatibility of any decision made around geological data, 725 especially for terranes or blocks that have limited palaeomagnetic data. For example, if the data support 726 two or three interpretations in the early Neoproterozoic, but only one of those is consistent with an 727 Ediacaran (or younger position), then we consider that position more reliable than the other two. However, 728 we argue that this logic also works in reverse; if a number of positions are deemed viable in the Ediacaran 729 for a terrane based on the data available, but only one of those also fits what data are available in the early 730 Neoproterozoic, then we will use the older data to force an interpretation of the younger data (e.g. Evans, 731 2009). This argument is highlighted by the concept of ‘world uncertainty’ (the percentage of total crust 732 (oceanic and continental) on the earth at any one time that is also preserved at present-day) of Torsvik et 733 al. (2010b). Because time is asymmetrical, this means that the level of confidence we have at present-day 734 is 100%, but decreases linearly back in time to ~70% at 200 Ma (i.e. 70% of all crust at 200 Ma is no longer 735 preserved). At 400 Ma, the world certainty is ~73% (Domeier and Torsvik, 2017), and using estimates of 736 continental crustal volumes, at 600 Ma it is between 75 and 80% (Cawood et al., 2013a). Therefore, for the 737 Mesozoic and Cenozoic, data from younger times are much more compelling to force interpretations of 738 older data because of how much more confident one can be in the last 20 Ma. Comparably, the difference 739 in this measure of uncertainty is much smaller between the Neoproterozoic and Cambrian, thus we suggest 740 that models in this time period should simultaneously use older and younger data to iterate towards a 741 solution. 742 743 4 Palaeomagnetic Data 744 745 A major problem in comparing the Phanerozoic models is that they all use different rotational frameworks, 746 including reference trees (i.e. flat vs hierarchical) and different absolute reference frames. In order to 747 properly synthesise the DOM16, DOM18 and YOU19 models into a single reconstruction, we first need to 748 lay a coherent groundwork in defining an absolute plate motion model for the largest continents during this 749 time in order to merge the individual models. To do this, we first derive a new APWP for Gondwana (540– 750 320 Ma) and GAPWaP for Pangea (320–0 Ma) using the palaeomagnetic compilation of Torsvik et al. 751 (2012) to provide an absolute palaeomagnetic reference frame for 540 to 0 Ma. We also apply the GAPWaP 752 to the MUL16 portion (i.e. the Mesozoic and Cenozoic) of the YOU19 model. We do this to ensure 753 compatibility between the Cenozoic and Palaeozoic, and also because our goals for this model are broad 754 scale (> 10–100 Ma) trends, mostly focussed on the Neoproterozoic. The method we follow to calculate 755 our APWP and GAPWaP is outlined below. For the Neoproterozoic, and non-Gondwana constituents of 756 the Palaeozoic, we use the compilations of palaeomagnetic data as presented in MER17, DOM16 and 757 DOM18, along with some other additions. Palaeomagnetically derived alterations to the models are also 758 discussed below. 759 760 4.1 APWP and GAPWaP construction 761 762 The absolute reference frames for Gondwana (540–320 Ma) and Pangea (320–0 Ma) used in this study are 763 derived using the method and velocity-optimised global palaeomagnetic data of Tetley (2018). APWPs are 764 routinely constructed using poles assigned averaged or nominal ages, which particularly for older times, 765 where palaeomagnetic constraint becomes increasingly limited, contribute to spurious apparent polar 766 wander behaviours. This method directly evaluate individual palaeomagnetic pole age and associated 767 uncertainty in combination with calculated pole A95 latitude and longitude uncertainties to derive optimised 768 APWPs that minimise predicted plate velocities and plate velocity gradients (instantaneous accelerations). 769 The resulting rate of apparent polar wander in optimised APWPs was reduced globally by an average of 770 56% by comparison to existing APWPs, resulting in predicted Phanerozoic plate motions displaying greater 771 kinematic consistency with present-day plate motion behaviours. 772 773 Optimised APWPs were produced for the 15 major continental blocks of Amazonia, Australia, Colorado, 774 East Antarctica, Greenland, India-Pakistan, Madagascar, Meseta, North America, Northeast Africa, 775 Northwest Africa, Panama, , Somalia, and Stable Europe. Applying the method as described in 776 Torsvik et al. (2012, 2008) and the data provided in Torsvik et a. (2012), optimised continental pole data 777 from all 15 continents were rotated from their individual source coordinate frames into a consistent South 778 African coordinate frame using the rotation model from this study. Now in a consistent coordinate frame, 779 a GAPWaP for Pangea was produced using all poles aged 320-0 Ma (due to the collision of Laurussia and 780 Gondwana during the Late Carboniferous), with a second GAPWaP produced using poles aged 540-320 781 Ma associated with Gondwana (South America, Africa, India, Antarctica and Australia) only. For both 782 GAPWaP reference frames, the GMAP software was used (Torsvik et al., 2012, 2008; Torsvik and 783 Smethurst, 1989), applying a running mean method using a window size of 20 Ma and a step size of 10 Ma. 784 785 4.2 Palaeomagnetic compilation 786 787 A compilation of palaeomagnetic data was used to constrain the position of all continental blocks during 788 the Neoproterozoic and for non-Gondwanan continental blocks during the early and middle Palaeozoic. 789 The Neoproterozoic data are presented in Table 2 and a GPlates compatible file of the data is also presented 790 in the Supplementary Material. Figure 6a–c shows the great circle misfit of our model to the selected poles. 791 As the majority of the data have already been used in the MER17, DOM16 and DOM18 reconstructions, 792 we point readers to those publications for in-depth discussion of the data. Here, we discuss two alterations 793 to the base models that we implemented based on palaeomagnetic data: Tarim at ca. 700 Ma and Laurussia 794 at ca. 420–400 Ma. 795 796 The MER17 model omitted Tarim prior to 700 Ma due to its pre-700 Ma palaeomagnetic data nullifying 797 the position that placed it outboard of Australia. We rectify this and include a robust, time-sensitive 798 geological argument for its position against India-South China during the Tonian (1000–720 Ma) and 799 Cryogenian (720–635 Ma) (Section 5.4.2). The rationale for placing Tarim in this position is to allow for a 800 significant 180° rotation required by paleomagnetic data. Two well-dated and high-quality poles; the 760– 801 720 Ma Qiaoenbrak Formation (Wen et al., 2013) and the 770–717 Ma Baiyisi Volcanics (Huang et al., 802 2005) are internally consistent and require Tarim to be inverted 180° from its present-day position (i.e. 803 northern margin facing south). A third pole, assumed to be pre 700 Ma, from the Aksu Dykes (Chen et al., 804 2004) is rejected for poor age constraints and the possibility of remagnetisation (Wen et al., 2017). 805 Comparably, three younger poles: the 635–550 Ma Sugetbrak Formation (Zhan et al., 2007), the ca. 635 806 Ma Tereeken Cap Carbonate (Zhao et al., 2014) and the 621–609 Ma Zhamoketi Andesite (Zhao et al., 807 2014) are all consistent with each other and indicate that Tarim was in its present-day orientation. A recent 808 pole by Wen et al. (2017) from the Lower Sugetbrak formation (640–615 Ma) stands in conflict with these 809 three poles, suggesting a ~50° rotational difference in the orientation of Tarim, while maintaining the same 810 palaeolatitude. Wen et al. (2017) dismiss the three earlier poles due to similarity to –Devonian poles 811 for the Sugetbrak Formation and possible remagnetisation for the latter two poles, respectively. 812 813 Successfully fitting the two older poles with either the three younger poles or the single pole of Wen et al. 814 (2017) is not possible in a model in which Tarim is surrounded by continental lithosphere for the 815 Neoproterozoic or in a scenario where Tarim is attached to the north-western or northern margin of 816 Australia or the northern margin of India (e.g. instead of South China). An accommodation of the data can 817 be obtained by placing Tarim in a ‘Missing-Link’ position (between Australia and Laurentia) with breakout 818 and rotation of Tarim from ca. 700 Ma; as argued by Wen et al. (2018, 2017). However, beyond the 819 kinematic issues with Missing-Link style models (Merdith et al., 2017b), it is difficult to account for the 820 formation of the 760 Ma Aksu Blueschist if Tarim was located in the centre of an assembled Rodinia (see 821 comment by Song and Li (2019) and reply by Wen et al. (2019)). In the present model, we use the pole of 822 Wen et al. (2017) due to its greater reliability, though our model could easily be adapted to fit the other 823 three poles if necessary (Section 5.4.2.). We also note that although we argue that India-South China were 824 separate from the rest of Rodinia, our preferred Neoproterozoic position for Tarim is also compatible with 825 models in which India-South China formed part of Rodinia (e.g. Cawood et al., 2013b). 826 827 In the YOU19 model, Laurussia was moved ~30° further north in latitude relative to its position in DT14 828 and MAT16 (Section 3.3). The consequence of this decision by YOU19 is that Laurussia is not in a 829 palaeomagnetically constrained position in the late Silurian and early Devonian. For this model, in addition 830 to the compilation of palaeomagnetic data in Torsvik et al. (2012), we also used palaeomagnetic data from 831 mainland Baltica (Table 2) to constrain the late Silurian to early Devonian position of Laurussia. We 832 implement these alterations to produce a coherent motion of Laurentia (later Laurussia) and Gondwana that 833 fits palaeomagnetic data, while also ensuring that the relative motion between Gondwana and Laurussia is 834 convergent, rather than dextral transform (e.g. Fig. 2c). 835 836

Figure 6 Summary of the fit of palaeomagnetic data to our model for 1000 to 400 Ma (omitting poles used to construct the Gondwana APWP). Misfit is the minimum great circle distance (in degrees) within the valid time range between the palaeomagnetic pole and the geographic north pole. The error bar on each point is the pole 95% confidence limit (A95). Solid line is the mean misfit of all poles, with the dashed lines representing the standard deviation of the mean. Poles marked as ‘not used’ or ‘inclination only’ in Table 1 are not included in this figure. (a) Poles from the constituent cratons of western Gondwana and the Avalonian terranes; (b) poles from the constituent cratons of eastern Gondwana and present-day Asia, and; (c) poles from Laurussia and Siberia. 837 838 5 Alterations to models 839 840 We describe the alterations made to individual models separately for clarity, however, we stress that no 841 single model was treated in isolation. Each alteration, whether during the Tonian or Devonian, was 842 evaluated both forwards and backwards in time in order to ensure continuity with both older and younger 843 geological and/or palaeomagnetic evidence. We begin our discussion with the alterations made to YOU19, 844 followed by alterations to DOM16, DOM18 and MER17. The alterations to DOM16 and DOM18 are 845 discussed together, as the two models essentially form a single global model between 500 and 410 Ma. The 846 most significant changes, and the focus of much of the discussion, occurred in MER17. This is because, 847 firstly, connecting this model with younger models in order to validate Neoproterozoic tectonic-geography 848 is a primary goal of this study, and secondly, because many of the alterations are completely new and have 849 never been incorporated into a plate model before. Figure 7a–b displays a latitudinal overview of the major 850 and minor cratons in our model, with a comparison to the DOM-MAT models for the Phanerozoic, and the 851 model of Li et al. (2013, 2008) for the Neoproterozoic. Figures 8, 9 and 10 show overviews of our 852 reconstruction at 1000 Ma, 500 Ma and present-day with relevant terranes and blocks highlighted. 853

Figure 7 Comparison of palaeolatitude of major cratons (a) and Chinese cratons (b) from 1000–0 Ma between three models; (i) This model, M2020; (ii) DOM16/18 (500–410 Ma) and MAT16 (410–0 Ma), and; (iii) Li et al. (2013, 2008) (1000–550 Ma). The notable excursion at 800 Ma in the Li et al. (2013, 2008) model is due to their adoption of IITPW at this time (Li et al., 2004). 854 855

Figure 8 (a) Distribution of continental crust, ocean basins and plate boundaries in our plate model at 0 Ma, based on MUL16. Tan polygons are areas of continental lithosphere in the Neoproterozoic that we model, blue polygons are areas of present-day continental lithosphere. Abbreviations as Figures 8 and 9. A, Amazonia; Ant, Antarctica; AUS, Australia; BP, Borborema Province; CCT, Chinese Central Tianshan; Cu, ; CY, Chu Yili; K, Kalahari; KMT, Krygyz Middle Tianshan; KNT, Krygyz North Tianshan; N-B, Nigeria-Benin; NC, North China; P, Paranapanema; ANS, Arabian Nubian ; Qa, Qaidam; Qi, Qilian; RDLP, Rio de la Plata; SC, South China; SM, Sahara Metacraton; T, Tarim. 856 857

Figure 9 Distribution of continental crust, ocean basins and plate boundaries in this plate model at 500 Ma, based on DOM16 and DOM18 (e.g. Figs. 12 and 13). Tan polygons are areas of continental lithosphere in the Neoproterozoic that we model, blue polygons are areas of present-day continental lithosphere that are inferred to exist during the Neoproterozoic, but without having firm geological evidence or that have been effected by subsequent deformation (e.g. the distance between a present-day coastline and COB). Green polygons represent a schematic interpretation of congruent continental lithosphere, with intervening crust being subsequently deformed during future tectonic cycles. Abbreviations as Figure 8, in addition to; Af, Afghanistan; B, Baltica; C, Congo; H, Hoggar; I, India; S, Siberia. 858 859

Figure 10 Distribution of continental crust, ocean basins and plate boundaries in our plate model at 1 Ga. Tan polygons are areas of continental lithosphere in the Neoproterozoic that we model, blue polygons are areas of present-day continental lithosphere that are inferred to exist during the Neoproterozoic, but without having firm geological evidence or that have been affected by subsequent deformation (e.g. the distance between a present-day coastline and COB). Green polygons represent a schematic interpretation of congruent continental lithosphere, with intervening crust being subsequently deformed during future tectonic cycles. A-Ant, Austral-Antarctica; Az, Azania; BP, Borborema Province; CLB, Coats Land Block; Dabo. Arc, Dabolava Arc; DML, Dronning Maud Land; eSTR, Eastern South Tasman Rise; IOGU, In Ouzal Unit; IGU, Iforas Granulite Unit; KMT(I), Krygyz Middle Tianshan (Issedonian); KMT(UM), Krygyz Middle Tianshan (Ulutau-Moyunkum); KNT(I), Krygyz North Tianshan (Issedonian); Mad, Madagascar; NAC, North Australian Craton; NeoMoz. Ocean, NeoMozambique Ocean; NC, North China; p-ANS, proto-Arabian Nubian Shield; Ra, Rayner Province; Ru, Ruker Block; SAC, South Australian Craton; SB, Southern Borborema; SL, ; Tas, Tasmania; wSTR, Western South Tasman Rise. 860 861 5.1 YOU19 862 The primary change we made to YOU19 was the implementation of a new APWP for Gondwana from 540 863 to 320 Ma and a GAPWaP for 320 to 0 Ma (Tetley, 2018). We also adjusted the position of Laurussia 864 during the Devonian in order to better fit palaeomagnetic data and implemented an alternative position for 865 Lhasa (against northwest Australia) that we consider to be more consistent with geological data in the 866 Neoproterozoic (Section 5.4.1). 867 868 In the YOU19 reconstruction, Laurussia is rotated roughly 40° counter-clockwise compared to the DT14 869 and MAT16 reconstructions (Fig. 11). While these changes improved the global kinematic integrity of the 870 model (Young et al., 2019), the position of Baltica in this configuration conflicts with palaeomagnetic data, 871 which indicates it lay at more southerly latitudes (Torsvik et al., 2012). We used the compilation of Torsvik 872 et al. (2012) as a base, however, as many of the poles in that dataset for the late Silurian to early Devonian 873 were taken from the (c.f. Jeleńska et al., 2015), we supplement it with some data from cratonic, 874 continental Europe (Table 2, S. Pisarevsky pers. comm.). We modify the position of Laurussia from that 875 used in YOU19 such that it fits the cluster of poles at this time (Fig. 11b). This modification places it in a 876 similar latitudinal position to DT14 and MAT16 (e.g. Fig. 7a), however, we shift it further east relative to 877 DT14, so that it is closer to its final position in Pangea. This position then satisfies the kinematic and 878 structural issues outlined in YOU19, while maintaining Baltica and Laurussia at a palaeolatitude permitted 879 by palaeomagnetic data. 880 881

Figure 11 Comparison of our model (a, b) with Domeier (2018, 2016) (c, d) at 411–410 Ma to highlight the latitudinal changes implemented for Laurussia in our model. Dark outline of Laurussia and Lhasa (italicised labels) in (a) are their original positions in YOU19. Palaeomagnetic data are shown at time of best fit for both models (i.e. map projection view of Fig. 6). Tan polygons are areas of continental lithosphere in the Neoproterozoic that we model, blue polygons are areas of present-day continental lithosphere that are inferred to exist during the Neoproterozoic, but without having firm geological evidence or that have been affected by subsequent deformation (e.g. the distance between a present-day coastline and COB). Green polygons represent a schematic interpretation of congruent continental lithosphere, with intervening crust being subsequently deformed during future tectonic cycles. The DOM16/18 models extend until 410.1 Ma (so displayed at 411 Ma), while YOU19 begins at 410 Ma. NC, North China; SC, South China; T-K-Qi-Qa, Tarim-Kunlun-Qilian-Qaidam. 882 883 The Lhasa block, currently preserved in the Tibetan between India and Tarim, is an E-W elongated 884 block consisting of Precambrian metamorphic basement, overlain by predominantly Palaeozoic 885 sedimentary rocks and Mesozoic and Cenozoic volcanic assemblages (Yin and Harrison, 2000; Zhu et al., 886 2013). The Precambrian basement is established only in the southern and central terranes of Lhasa (Zhu et 887 al., 2013) and in the Amdo micro-block that is preserved in the Mesozoic northern (e.g. Fig. 888 10b). Lhasa is typically placed outboard of the Tethyan Himalayan terranes, off the northern margin of 889 India within Gondwana (e.g. Domeier and Torsvik, 2014). However, an alternative position outboard of 890 northwest Australia is also supported; a scenario consistent with tectonic affinities interpreted from zircon 891 age spectra and Hf isotopic signatures (Burrett et al., 2014; Zhu et al., 2011). We find a position off 892 northwest Australia more consistent with the Neoproterozoic geological record of Lhasa (Section 5.4.1) 893 that preserves magmatism interpreted to represent the existence of a subduction zone and back-arc (Guynn 894 et al., 2012; Hu et al., 2018), which would otherwise be impossible to produce if it was land-locked between 895 India and South China. We therefore alter the position of Lhasa from YOU19 to outboard of NW Australia. 896 For simplicity, we infer Lhasa’s motion to follow the drift of the Cimmerian terranes, but recognise that the 897 palaeolatitudes need further refinement in the Jurassic and Cretaceous following syntheses such as Li et al. 898 (2017, 2016) and others. 899 900 5.2 DOM16 and DOM18 901 902 We sought to preserve, as closely as possible, the palaeolatitudes and the tectonic interpretations (i.e. history 903 of collisions, rifting, subduction and ocean basin evolution) of the DOM16 and DOM18 reconstructions. 904 The position of Gondwana (both in Early Palaeozoic and the older times covered by MER17) has been 905 adjusted to inherit the Gondwana position at 410 Ma from the new Gondwana APWP path described above. 906 Relative rotations of smaller blocks to Gondwana were calculated from DOM16/18 and translated into the 907 new position of Gondwana. Further adjustments to all continental polygons have been implemented to 908 smooth the transition from the late Palaeozoic configurations inherited from the adjusted YOU19 model 909 (adj-YOU19). The following sections detail adjustments to the DOM16/18 models made for the late 910 Cambrian–Devonian. 911 912 Figure 11 shows a direct comparison between the DOM16/18 reconstructions and our adj-YOU19 model 913 at 411/410 Ma. The primary differences between the models are longitudinal, as our adj-YOU19 model 914 shifts Laurussia, Siberia, Kazak, Tairm and North China ~30° longitudinally to the east compared to 915 DOM16/18 (as well as when compared to DT14 and MAT16, which link closely to DOM16/18), in order 916 to better model the amalgamation of Laurussia and Gondwana. This results in a far narrower 917 at 410 Ma in our model and a much simpler collision between Laurussia and Gondwana (e.g. Wu et al., 918 2020; Young et al., 2019). This change is discussed further in context in the following section. 919 920 The motions of Baltica, Laurentia, and Siberia are tied directly to the spin axis through their own individual 921 palaeomagnetic reference frame, in contrast to the younger parts of the reconstruction where we retain the 922 hierarchy inherited from YOU19. For minor cratons (Tarim, North China, South China) and smaller blocks 923 (e.g. Tianshan, Kara), we model their motion within a relative hierarchy for the practical reason that this 924 makes it easier to preserve the consistency of their geological history, as they share multiple plate 925 boundaries with Gondwana. For all these cratons and blocks, we sought to preserve the palaeolatitudes from 926 the parent studies to a reasonable degree allowing for data uncertainties (e.g. Fig. 6). Figure 7 provides a 927 quantitative comparison between the paleolatitudes of the DOM16/18 models and our incorporation of them 928 into our adjusted reconstruction. 929 930 5.2.1 Deviations from DOM16 931 932 The key deviation between DOM16 and our model is a difference in the orientation of Baltica in the late 933 Cambrian (Fig. 12a), where our model has Baltica rotated 90° counter clockwise relative to DOM16. As 934 the DOM16 model only begins at 500 Ma it does not have to explicitly consider the earlier Neoproterozoic 935 history of Baltica. In our opinion the position of Baltica in DOM16 is more congruent with an inverted 936 Baltica (relative to Laurentia) during the Neoproterozoic (Hartz and Torsvik, 2002), where the southern- 937 peri Urals are connected to Greenland. In comparison, MER17 connects Baltica to Rodinia through the 938 Sveconorwegian margin in an upright position relative to the present-day (e.g. Cawood et al., 2003; Dalziel, 939 1992; Weil et al., 1998). The inverted Neoproterozoic Baltica position of Hartz and Torsvik (2002) results 940 in simpler kinematic motions during the late Ediacaran and Early Palaeozoic, whereas the ‘traditional’ 941 Neoproterozoic coupling of Baltica-Laurentia requires a more complex motion path in order to fit 942 palaeomagnetic data. However, in our opinion the ‘upright’ coupling of Baltica and Laurentia is far more 943 geologically and palaeomagnetically robust (Cawood and Pisarevsky, 2006; c.f. Slagstad et al., 2019) in 944 the Neoproterozoic than the alternative, and as such we adopt this configuration at ca. 600 Ma (as in 945 MER17) during the opening of the Iapetus Ocean; necessitating a more complex kinematic evolution for 946 Baltica between 600 and 470 Ma (Fig. 12a ,b). Our reconstruction of Baltica is therefore quite different to 947 that of DOM16 during the late Cambrian and early as we have to implement a ~90° rotation 948 between 520 and 475 Ma of Baltica to fit the same series of palaeomagnetic data at 470 Ma as DOM16. 949 950 Baltic palaeomagnetic data compiled by Meert et al. (2014), Torsvik et al. (2012) and Domeier (2016) were 951 used to ensure its latitudinal position remained valid between 550 and 470 Ma (Table 2). Importantly, only 952 two poles—one from the Andarum (Torsvik and Rehnström, (2001) categorised as C-grade quality 953 by Meert (2014)) and the other from the Narva sediments, (Khramov and Iosifidi, (2009), categorised as B- 954 grade quality by Meert (2014))—are identified for Baltica between 530 and 480 Ma, both with a nominal 955 age of 500 Ma. Despite coeval ages, the poles are ~35° apart from one another and Meert (2014) identified 956 unresolved issues, specifically: few samples constraining the pole and a (possible) effect of inclination 957 shallowing (even after correction for inclination shallowing a ~25° mismatch remains). A strict fitting of 958 the Narva sediments pole would require Baltica to rotate by ~2°/Ma between 550 and 500 Ma; a situation 959 we consider implausible and likely reflecting underlying issues in the palaeomagnetic data. As such, our 960 model does not fit either pole explicitly, though our reconstructed position of Baltica is consistent with the 961 latitude suggested by the inclination data of both poles. By 460–440 Ma, our model closely resembles the 962 DOM16 model, with similar sized Rheic oceans, latitudinal positions and orientations of Gondwana, Baltica 963 and Laurentia (Fig. 12c, d). At 410 Ma, when DOM16 finishes, the only difference is the relative position 964 of and Gondwana due to differences in palaeolongitude (Fig. 2c; 11). In our model, Laurussia is 965 much closer to Gondwana resulting in a far narrower Rheic Ocean at 410 Ma than in DOM16/18, DT14 966 and MAT16 (e.g. Fig. 2c, see also Wu et al. (2020)). 967 968 The width of the Rheic ocean is poorly constrained (c.f. Dalziel and Dewey, 2019; Domeier, 2018; Wu et 969 al., 2020) especially because of palaeolongitudinal uncertainty. The problem is compounded by the fact 970 that there is only one reliable palaeomagnetic pole constraining our Gondwana APWP between 430 and 971 400 Ma (Aïr intrusives in , age at 410 Ma, (Hargraves et al., 1987)), meaning the early Devonian 972 portion of the APWP has large uncertainty. Because of this uncertainty, the methods used to create APWPs 973 tend to dampen the effect of this pole (which is true for our APWP). Nonetheless, even with the effect of 974 this pole being dampened, reconstructed palaeomagnetic data (without considering longitudinal constraints) 975 at 410 Ma allow placement of Laurussia and Gondwana to within a few thousand kilometres of one another 976 (e.g. Fig. 11a, see also Wu et al. (2020)). At first glance, this seems problematic, because one might expect 977 that the forces related to subduction zones modelled along the craton margins facing each other would draw 978 the cratons to each other. DOM16, DT14 and MAT16 (along with Torsvik et al., (2014)) solve this problem 979 by changing the palaeolongitude of Laurussia to 90° east relative to Gondwana, allowing for a much wider 980 ocean basin. This position is justified and necessitated in these models by their use of the PGZ method, 981 where Laurussia is reconstructed over the eastern arm of the present-day position of the Pacific LLSVP. 982 However, this position then requires > 8000 km of dextral motion between Laurussia and the Patagonian 983 margin of Gondwana from 400 and 340 Ma (e.g. Fig. 3c). As we do not adopt the PGZ method, our position 984 of Laurussia at 410 Ma relative to Gondwana is much closer to its final collision place and the resulting 985 relative motion between Laurussia and Gondwana between 410 and 340 Ma is of sub-orthogonal– 986 orthogonal collision along the southern Appalachian zone (Hopper et al., 2017), with some dextral 987 transform motion between north-east Laurentia and the northwest margin of Gondwana (e.g. Murphy et al., 988 2011) and between southern Baltica and north Gondwana (e.g. Arthaud and Matte, 1977). From 340 to 320 989 Ma YOU19, MAT16 and DT14 have a similar configuration between Laurussia and Gondwana. We stress 990 that while our model this differs from the adopted DOM16 model, it is not a particularly novel interpretation 991 and many continental reconstructions show a similar scenario (e.g. McKerrow et al., 2000; Scotese, 2004; 992 Stampfli and Borel, 2002). A more detailed geological and kinematic justification of this interpretation is 993 presented in Young et al. (2019) and a discussion of the palaeomagnetic challenges in amalgamating Pangea 994 are presented in Domeier et al. (2020, 2012) which we encourage interested readers to. 995 996

Figure 12 Comparison of the evolution of the Rheic Ocean between our model (a, 500 Ma; c, 440 Ma) and DOM16 (b, 500 Ma; d, 440 Ma) during the early Palaeozoic. We point out the rotational difference (but similar latitude) in Baltica at 500 Ma in our model relative to DOM16, discussed in text. Palaeolongitude is constrained absolutely in DOM16 but not in our model. Subduction polarities are not provided in the GPlates files of DOM16. Tan polygons are areas of continental lithosphere in the Neoproterozoic that we model, blue polygons are areas of present-day continental lithosphere that are inferred to exist during the Neoproterozoic, but without having firm geological evidence or that have been affected by subsequent deformation (e.g. the distance between a present-day coastline and COB). Green polygons represent a schematic interpretation of congruent continental lithosphere, with intervening crust being subsequently deformed during future tectonic cycles. IO, Iapetus Ocean. 997 998 5.2.2 Deviations from DOM18 999 1000 Two salient points made by Domeier (2018) pertaining to relative longitude and his model are also of 1001 interest here, and worth reiterating when merging DOM18 into the adj-YOU19 model and then extending 1002 the adj-YOU19 model into the Neoproterozoic. Firstly, a long lived south-dipping subduction zone, 1003 preserved in the northern Kazakhstan terranes of Urumbai, Selety and Erementau (Degtyarev, 2011; 1004 Domeier, 2018; Windley et al., 2007) was longitudinally distributed from Siberia through to the 1005 northernmost margin of Gondwana (i.e. North Australia, Papua New Guinea) and secondly, the broader 1006 framework of Baltica-Siberia-Gondwana provides geological, spatial and temporal limits on the evolution 1007 of this area. These two aspects of the model allowed DOM18 to infer with some certainty the relative 1008 palaeolongitude of many of the smaller terranes within the broader absolute palaeolongitudinal constraints 1009 imposed by (principally) Siberia, Baltica, Laurentia and Gondwana. Likewise, even though we do not adopt 1010 the absolute longitudinal framework of DOM18, we can use the positions of the major cratons to provide a 1011 relative control on the possible kinematic evolution of the Chinese cratons and terranes. We follow DOM18 1012 by reconstructing a quasi-stable south-dipping subduction zone to delimit the northerly extent of the Palaeo- 1013 and this, coupled with the palaeomagnetic data from these cratons and terranes between 500 1014 and 410 Ma, make longitudinal re-ordering of these blocks unlikely. The unlikeliness of longitudinal re- 1015 ordering is one of the main constraints and pieces of evidence that to significant revision of the 1016 positions of Tarim and North China in MER17 that are discussed in the following section (5.4). 1017 1018 The relative positions of terranes (Tianshan, Qaidam-Qilian, Kunlun etc.) can also be considered in a similar 1019 manner to how we conceive of the relative ordering of the Chinese cratons during the Early Palaeozoic. In 1020 particular, the ca. 470 Ma suturing of Tianshan-Chu Yili (Alexeiev et al., 2019) and the 440–430 Ma 1021 suturing of Kunlun, Qaidam, Qilian and Alxa to Tarim (Xiao et al., 2009) both necessitate an internally 1022 consistent relative position between these terranes and Tarim and Siberia in order to be consistent with the 1023 geological record. That is, relative to present-day Tarim, Tianshan, Chu-Yili and other Kazakh terranes 1024 should be reconstructed somewhere north of the northern margin of Tarim during the Ediacaran– 1025 Ordovician, but south of Siberia. Similarly, Kunlun, Qaidam, Qilian and Alxa need to be reconstructed to 1026 the south and east of Tarim in the same time period. In a plate model framework, this means that the 1027 longitudinal structure of the Proto-Tethyan ocean basin should reconstruct (from west to east): Siberia— 1028 Chu-Yili-Kazak-Tianshan—Tarim—Alxa-Qaidam-Qilian-Kunlun—North China—South China (e.g. Fig. 1029 13f, g). Further back in time, we also maintain this same configuration to minimise any terrane re- 1030 organising, such that their relative positioning is broadly reminiscent of present-day (e.g. Figs. 9; 16). In 1031 this manner, we use these relative longitudinal constraints to infer a configuration for the nuclei of these 1032 terranes in the Neoproterozoic, thereby connecting the present-day with their Precambrian history. 1033 1034 The fundamental difference between DOM18 and our implementation of the model is the longitudinal width 1035 of the Proto-Tethyan ocean basin, bounded by Siberia in the west, Gondwana in the east and south and the 1036 aforementioned south-dipping subduction zone in the north (e.g. Fig. 13b, c, f, g). The latitudinal extent of 1037 the ocean basin remains similar in both models (30–50°), constrained by palaeomagnetic data from 1038 Gondwana, Siberia and the Chinese cratons (e.g. Table 2, Fig. 13a, d, e, h). In our model, this ocean basin 1039 is much wider (longitudinally) in the late Cambrian and Ordovician than in DOM18, narrowing in size as 1040 it evolves due to our implemented easterly drift of Siberia from the Cambrian through to the Devonian. The 1041 size of the ocean basin is then similar at 410 Ma (Fig. 11), due to the adopted similarity of the YOU19 from 1042 the DT14 reconstruction (Young et al., 2019). The key reason for the difference in width at 500 Ma is 1043 because DOM18 places Laurussia further east at 500 Ma than we do (Section 5.2.1). This then forces a 1044 narrower ocean basin between 500 and 410 Ma in DOM18 than in our model. The following paragraphs 1045 will discuss the regional longitudinal constraints of this ocean basin by considering the position of 1046 Gondwana and Siberia. 1047 1048 The Early Palaeozoic position of Siberia in our model is a function of palaeomagnetic data and its position 1049 in, and breakout from, Rodinia. In isolation, the simplest explanation of Siberia’s journey is that sometime 1050 during the late Tonian–Cryogenian (750–700 Ma) Siberia rifted off the northern margin of Laurentia 1051 (somewhere near Greenland, see Pisarevsky and Natapov (2003) and Pisarevsky et al. (2013)). At this time 1052 Siberia was located equatorially and rotated 60° clockwise from its present-day orientation (Pisarevsky et 1053 al., 2013). Palaeomagnetic data are sparse for the remainder of the Neoproterozoic, with the few calculated 1054 poles having either poor age constraints or unresolved tectonic coherence with the Siberian craton (Pavlov 1055 et al., 2015), and are therefore typically omitted from syntheses (Cocks and Torsvik, 2007; Li et al., 2008; 1056 Merdith et al., 2017a). However, from the mid-Cambrian the palaeomagnetic record of Siberia is reasonable 1057 (Cocks and Torsvik, 2007) and broadly congruent with the palaeolatitude of the 750 Ma pole (that is, 1058 equatorial–sub-equatorial). From the mid-Cambrian, the data suggest a slow northward drift and counter- 1059 clockwise rotation (Cocks and Torsvik, 2007), with the orientation of Siberia at ca. 530 Ma inverted relative 1060 to present-day. This then requires a ~120° clockwise rotation between 720 and 530 Ma in order to fit its 1061 Neoproterozoic position (Metelkin et al., 2012). For the Cryogenian and Cambrian, Siberia's motion can be 1062 inferred indirectly using data from Baltica, as outlined in the next paragraph. 1063 1064 The position of Baltica is constrained by clusters of palaeomagnetic data during the Ediacaran and early 1065 Cambrian. Furthermore, its latitudinal position places limits on the possible position of Siberia. The 1066 equatorial excursion of Baltica in the latest Ediacaran places Baltica at a latitude similar to Siberia. As they 1067 cannot be positioned on the same longitude (Merdith et al., 2017a), Siberia must be lcoated either 1068 longitudinally east or west of Baltica between 600 and 500 Ma (with Laurentia also occurring on a similar 1069 palaeolatitude but further east than either Siberia or Baltica). Domeier (2018) presented similar arguments 1070 for the second latitudinal excursion of Baltica (it returns to a high latitude during the Ordovician–Silurian) 1071 when it collided with Laurentia to form Laurussia at ca. 440–430 Ma. He argued that the overlapping 1072 palaeolatitudes of Siberia and Baltica at this time then requires Siberia to be located more easterly than 1073 Baltica by ca. 470 Ma (Domeier, 2018). We also adopt this logic, thus providing a rough relative 1074 longitudinal framework for Laurentia-Baltica-Siberia relations from 700 to 450 Ma. 1075 1076 Our model requires the motion of Siberia to be predominantly longitudinal between 700 and 450 Ma (the 1077 available palaeomagnetic data do not suggest more than 30° latitudinal movement). We constrain this to 1078 two broad phases of movement, defined by Baltica’s two latitudinal excursions (at ca. 560 Ma and 450 Ma). 1079 In the first excursion (750–550 Ma), we keep Siberia longitudinally between Laurentia and Baltica, because 1080 to move it further east than Baltica at 550 Ma would require relative plate motion greater than 30 cm/a, 1081 which we deem unlikely. By 470 Ma palaeomagnetic data from Baltica suggest it has started drifting north 1082 again, so we therefore reconstruct Siberia to also be moving east longitudinally between 550 and 470 Ma, 1083 such that by 470 Ma it is located along the same longitude as Baltica; by 450 Ma it is further east than 1084 Baltica, such that Laurussia can form by 430 Ma. Our model is therefore similar to that of DOM18 in 1085 concept and adherence to available observations, however the different absolute longitudinal positions 1086 create a notably different ocean basin in the late Cambrian and Ordovician, within which the Chinese 1087 cratons and terranes are then arranged. 1088 1089 Domeier (2018) does not explicitly consider the Neoproterozoic or Early Palaeozoic evolution of the 1090 Chinese cratons in his model. At 500 Ma, the DOM18 model places North China off the northern margin 1091 of India, and South China off northern Australia. However, retaining these relationships in the 1092 Neoproterozoic is invalidated by what Neoproterozoic palaeomagnetic data exists for each craton (e.g. Fu 1093 et al., 2015; Li et al., 2004), and also conflicts with interpreted geological histories (Cawood et al., 2018b, 1094 2013b). Consequently, our model, which balances both older and younger times, alters the kinematic 1095 evolution in order to fit older constraints (Fig. 13, Sections 5.4.1 and 5.4.2). 1096 1097

Figure 13 Comparison of our model (a, c) with DOM18 (b, d) to highlight changes made. Annamia (Indochina, Sibumasu) is not explicitly modelled in our reconstruction. L, Lut (Iran); F, Farah (Afghanistan); T, Tarim. Tan polygons are areas of continental lithosphere in the Neoproterozoic that we model, blue polygons are areas of present-day continental lithosphere that are inferred to exist during the Neoproterozoic, but without having firm geological evidence or that have been affected by subsequent deformation (e.g. the distance between a present-day coastline and COB). Green polygons represent a schematic interpretation of congruent continental lithosphere, with intervening crust being subsequently deformed during future tectonic cycles. Al, Alxa; An, Annamia; A-S, Altai-Sinai; SC, South China; K, Kunlun; KMT, Krygyz Middle Tianshan; KNT, Krygyz North Tianshan; NC, North China; Qa-Qi, Qaidam-Qilian; Ql, Qinling; T, Tarim. 1098 1099 5.4 MER17 1100 5.4.1 Australia, North China, Lhasa and Tasmania 1101 Significant alterations to MER17 were made between 1000 and 900 Ma along the north-western, western 1102 and south-western margins of Laurentia, affecting the motions and positions of Australia, North China, 1103 Siberia, Tasmania and Lhasa. We adopt the model of Wen et al. (2018) in having a dextral shear zone 1104 between Australia-Antarctica (A-A) and Laurentia during the early Tonian. Wen et al. (2018) argued for 1105 placing Tarim against the eastern margin of Laurentia, separating Laurentia from Australia in a ‘Missing 1106 Link’ position (cf. (Li et al., 2008, 1995), with the dextral shear zone transecting Tarim. However, we 1107 consider this position for Tarim is incompatible both with geological data (the 760 Ma Aksu Blueschist 1108 (C.-L. Zhang et al., 2013)) and with the kinematic constraints that would be necessary to move Tarim from 1109 this position to its Palaeozoic position (e.g. Merdith et al., 2017b). We therefore use the alteration of Wen 1110 et al.’s (2018) model presented in Mulder et al. (2018b), who place a dextral boundary separating the 1111 Antarctic crust of Australian affinity exposed in Terre Adélie Land from the Antarctic crust of Laurentian 1112 affinity in the Nimrod Igneous Province (Fig. 14a, see also Goodge et al., (2017)). 1113 1114 North China and northern Australia share a similar and early Tonian sedimentary record 1115 and both preserve contemporaneous ca. 1.33–1.31 Ga magmatism (Bodorkos et al., 2020; Yang et al., 2020; 1116 Zhang et al., 2017) that is interpreted as a . Yang et al. (2019) also demonstrated the 1117 similarity in detrital zircon ages and isotope compositions between the Tonian strata of both areas. 1118 We find that this position of North China in the latest Mesoproterozoic is remarkably compatible with the 1119 few reliable palaeomagnetic data for North China in the Neoproterozoic (e.g. Fig 6, (e.g. Fig. 6, Fu et al., 1120 2015) and places North China in a position readily compatible with its Palaeozoic constraints where the 1121 same species of distinctive tommotiid fossils have recently been reported (Pan et al., 2018). A distinct Sino- 1122 Australian Cambro–Ordovician faunal province was identified by Burrett et al. (1990) that suggests some 1123 proximity in the early Palaeozoic. Cambrian–Ordovician rifting in the Arafura Basin north of northern 1124 Australia may represent the initial separation of North China from this margin of Gondwana (Ahmad and 1125 Munson, 2013). Palaeomagnetic data necessitate some relative motion between Australia and North China 1126 in the Early Palaeozoic from its inferred Mesoproterozoic–Neoproterozoic position (e.g. Domeier, 2018). 1127 Given there is no evidence of orogenesis between North China and northern Australia in the Phanerozoic, 1128 we infer that North China slowly drifts off this margin from the Cambrian, remaining in close enough 1129 proximity to share the identified Cambro–Ordovician faunal provinces (e.g. Fig. 13). 1130 1131 Dong and Santosh (2016) and Dong et al. (2014) describe a 1000 to 900 Ma between the Qinling 1132 Terrane and North China, preserved as the Kuanping (Fig. 14a–c). As Siberia and Australia are 1133 reconstructed adjacent to each other in Rodinia (Pisarevsky et al., 2013), the position of North China along 1134 the northern margin of Australia suggests that the Qinling terrane could feasibly be an extension of the 1135 Central Angara terrane, where there is a similarly aged (but sparsely described) ophiolite, the Ribnaya- 1136 Panimba ophiolite (Vernikovsky et al., 2004, 2003). In our model, the subduction zones represented by 1137 these two consume the oceanic lithosphere between Australia, Siberia and Laurentia (the 1138 Kuanping Ocean) during the early Tonian. Mulder et al. (2018b) ceased motion at 900 Ma in their model 1139 but, we adjust this cessation to 930 Ma wherein the Qinling Terrane rotates to fit against the North China 1140 block. This is because we also reconstruct the Lhasa block along the western margin of Australia ((Zhu et 1141 al., 2011), see Section 5.1) and here magmatism is preserved from ca. 925 Ma (Guynn et al., 2012, 2006; 1142 Hu et al., 2018; Zeng et al., 2018) (Fig. 14d,e). Consequently, we suggest that this subduction initiated after 1143 the closure of the interior Kuanping Ocean and collision of North China-Australia-Antarctica with Siberia- 1144 Laurentia along the Qinling-Central Angaran Terrane. The subduction zone then connects northwards 1145 through to subduction preserved in Taimyr outboard of Siberia (Vernikovsky et al., 2004; Vernikovsky and 1146 Vernikovskaya, 2001) and southward into an oceanic arc outboard of the Mawson Craton of Antarctica, 1147 possibly preserved in the southernmost Tonian Oceanic Arc Super Terrane (TOAST) or between Indo- 1148 Antarctica and Australia-Antarctica). In our model, Australia sits in a typical SWEAT (South West United 1149 States, East Antarctica) configuration (Moores, 1991). We made this change to better fit the arguments put 1150 forward by Mulder et al. (2018b), while still maintaining integrity of relative plate kinematics following 1151 the reasoning of Merdith et al. (2017b). 1152 1153 Our revised model also incorporates recent refinements to the Proterozoic and early Palaeozoic 1154 paleogeography of the Western Tasmania Terrane. The Western Tasmania Terrane, comprising the 1155 Proterozoic and the East and West South Tasman Rises (Berry et al., 2008), occupies 1156 an important position in deciphering the geological relationship between Laurentia and Australia-Antarctica 1157 in Rodinia, and also in understanding the transition between Rodinia to Gondwana. The Western Tasmania 1158 Terrane represents an exotic Proterozoic microcontinent that was accreted onto the Pacific margin of eastern 1159 Gondwana in the late Cambrian during the Ross-Delamerian orogenic cycle (Berry et al., 2008; Cayley, 1160 2011). The terrane has geological affinities with the central of East Antarctica 1161 and the western margin of Laurentia, including overlapping Palaeoproterozoic basement ages, 1162 contemporaneous Mesoproterozoic magmatic and fluid-flow events, and correlated Mesoproterozoic 1163 sedimentary strata (Berry et al., 2008; Fioretti et al., 2005; Halpin et al., 2014; Mulder et al., 2015, 2018b). 1164 Based on these geological connections, the Western Tasmania Terrane was likely located between East 1165 Antarctica and western Laurentia within an assembled Rodinia. The breakout of the Western Tasmania 1166 Terrane from its central position within Rodinia is recorded by widespread Tonian–Ediacaran 1167 sedimentation and rift-related magmatism (Mulder et al., 2020). The onset of Neoproterozoic rifting of the 1168 Western Tasmania Terrane is marked by 780–750 Ma intraplate magmatism (Black, 1997; Calver et al., 1169 2013) and latest Tonian (750–730 Ma) siliciclastic and carbonate sedimentation (Calver et al., 2014; Mulder 1170 et al., 2018a). Following deposition of Cryogenian rift-related strata and glaciogenic intervals (Calver, 1171 2011; Calver et al., 2014) a final pulse of Neoproterozoic rifting is recorded by voluminous ca. 580 Ma rift- 1172 related basalts in northwest Tasmania (Direen and Crawford, 2003; Meffre et al., 2004). Geological 1173 correlations permit the Western Tasmania Terrane to have remained attached to either the western margin 1174 of Laurentia or the eastern margin of Australia-Antarctica following the opening of the Pacific Ocean (Fig. 1175 10f), prior to being isolated as a microcontinent during ca. 580 Ma rifting and accretion onto its present- 1176 day position along the margin of Gondwana by the late Cambrian (Fig. 10g; Berry et al., 2008; Mulder et 1177 al., 2020). We follow Mulder et al. (2020) in having Tasmania rift from the Antarctic margin (rather than 1178 the alternative scenario of Laurentia), thus implying that some further unknown micro-continents rifted 1179 from the western margin of Laurentia at the same time in order to account for that (e.g. Fig. 1180 14f, g Macdonald et al., 2013) (see also Colpron et al., 2002; Cox et al., 2018; Eyster et al., 2019). 1181

Figure 14 Snapshots of plate reconstructions showing our updated model for Australia-Laurentia at key time intervals, along with palaeomagnetic data. (a) 1000 Ma; (b) 975 Ma; (c) 950 Ma; (e) 925 Ma; (f) 900 Ma; (h) 580 Ma and (i) 560 Ma. The Tonian model (a–f) follows arguments laid out in Mulder et al. (2018b), while the Ediacaran-Cambrian evolution is after Mulder et al. (2020). Times in these panels reflex the nominal time of best fit for each pole. Tan polygons are areas of continental lithosphere in the Neoproterozoic that we model, blue polygons are areas of present-day continental lithosphere that are inferred to exist during the Neoproterozoic, but without having firm geological evidence or that have been affected by subsequent deformation (e.g. the distance between a present-day coastline and COB). Green polygons represent a schematic interpretation of congruent continental lithosphere, with intervening crust being subsequently deformed during future tectonic cycles. CLB: Coats Land Block; DML, Dronning Maud Land; eSTR: Eastern South Tasman Rise; MO, Mirovoi Ocean; MS, ; Tas: Tasmania; wSTR: Western South Tasman Rise; RDLP: Río de la Plata; WAC, . 1182 1183 5.4.2 India-South China Accretionary Belt 1184 We preserve the MER17 interpretation of a tight India-South China connection (after Cawood et al., 2013b). 1185 This possible connection was suggested previously by Jiang et al. (2003) who noted the similarity between 1186 sequence stratigraphy in rift basins preserved in both South China and the Indian Lesser Himalaya (Fig. 1187 15a), as well as by Hofmann et al. (2011) who suggested a geological similarity based on detrital zircon 1188 analysis. Arguments for this connection are succinctly summarised in Cawood et al. (2018b) and are not 1189 repeated here—instead we focus our discussion of this margin on the relative position of outboard terranes 1190 during the Neoproterozoic (Fig. 15), which are based predominantly on their Palaeozoic positions 1191 (Domeier, 2018). Here we have sought to preserve their relative internal positions in order to minimise 1192 reshuffling of terranes during the Neoproterozoic (e.g. Fig. 16). For example, Kunlun, which is currently 1193 preserved south of Tarim and west of Qaidam-Qilian, is always reconstructed with the same internal 1194 consistency. Although we note that this may not be an accurate reflection of the ordering and positioning 1195 of the terranes, it ensures consistency within the model and minimises terrane shuffling which can preclude 1196 unrealistic scenarios, where terranes have to kinematically skirt one another precariously. Figure 16 gives 1197 a schematic overview of our conceptualisation and implementation of this model. 1198 1199 Following Alessio et al. (2018) and Armistead et al. (2019), the northwest margin of India is here interpreted 1200 as an extensive Stenian-Tonian accretionary margin that extends as far as the Omani basement and 1201 northernmost Madagascar. The pre-Ediacaran basement rocks in Rajasthan and Pakistan share similarities 1202 with those of Oman. Granitoids have been dated from Rajasthan and from Nagar Parkar in eastern Sind 1203 (Pakistan) at ca. 1.1 Ga (Meert et al., 2013; Raza et al., 2012). There is no evidence of older crust occurs 1204 west of the Western Margin of the Aravalli-Delhi Orogen, where the Marwar terrane accreted to India 1205 in the latest Mesoproterozoic (Meert et al., 2010) (Fig. 15b). Tonian granitoids and rhyolites occur in inliers 1206 through northwest India and Pakistan, where they cluster into crystallisation ages of ca. 990–970 Ma 1207 (Haldar and Deb, 2001; Pandit et al., 2003), ca. 860–820 Ma (Davies and Crawford, 1971; Deb et al., 2001; 1208 Just et al., 2011; Van Lente et al., 2009) and ca. 775–760 Ma (Ashwal et al., 2013; Gregory et al., 2009; 1209 Meert et al., 2013; Van Lente et al., 2009). The latter magmatic and extrusive phase forms one of the largest 1210 igneous provinces on the planet—the Malani Igneous Suite—which is also traced to the Seychelles 1211 (Fig. 15c–e) (Torsvik et al., 2001; Tucker et al., 2001). Arc accretion continued outboard to Oman where 1212 two main phases of subduction and arc magmatism occur, at ca. 850 Ma and ca. 770 Ma (Blades et al., 1213 2019a). The latter phase focussed in the southern Mirbat area and interpreted here as the arc that formed 1214 ocean-ward of the more back-arc Malani Igneous Suite. Further outboard still, and later accreting onto the 1215 Indian margin, the Bobakindro Terrane of northern Madagascar (Armistead et al., 2019) consists of juvenile 1216 magmatism that dates from ca. 750–705 Ma (Armistead et al., 2019; Collins, 2006; Thomas et al., 2009). 1217 1218 Many terranes currently preserved north of South China and south of Siberia have Neoproterozoic or older 1219 cores. They have not been previously considered in global models (Li et al., 2008; Merdith et al., 2017a) 1220 due to the sparsity of data and small size of terranes, which invites many competing and conflicting 1221 interpretations of their history. However, in the construction of this continuous plate model, where spatial 1222 and temporal continuity is vital, the most compatible Tonian position for these terranes was outboard of the 1223 afore-discussed large accretionary subduction zone of South China and India. This position places them in 1224 the most favourable kinematic, palaeomagnetic and geologically plausible positions for their (more well 1225 constrained) Palaeozoic journeys (e.g. Charvet et al., 2011; Domeier, 2018; Xiao et al., 2013). Below we 1226 summarise some geological evidence for this, with particular reference to the Tarim Craton, as it is the only 1227 block that has multiple reliable palaeomagnetic data from the Neoproterozoic that act as a another line of 1228 evidence. We also note that Huang et al. (2019) recently proposed a location outboard of Greenland for the 1229 Yili-Tianshan Block on the basis of similar detrital zircon age spectra. However, more work would have to 1230 be done to determine whether this Tonian position is consistent with the kinematic evolution necessary for 1231 these blocks to fit their Palaeozoic constraints. 1232

Figure 15 (a) Regional map at present day of Central Asian Orogenic Belt with modelled terranes highlighted in tan. Light tan terranes are not explicitly modelled but are referred to in the main text. Black areas represent the cratonic components. For the Kazakh area, we use polygons consistent with their Palaeozoic structure. The size, orientation and distribution of crust in these terranes in the Neoproterozoic is unknown due to the subsequent reworking of the terranes. Therefore, the precise position (along a margin), orientation, size and shape of these terrane polygons in the reconstruction figures are speculative and should be treated cautiously. (b–f) Evolution of the India-South China system during the Tonian–Cryogenian at key times. Tan polygons are areas of continental lithosphere in the Neoproterozoic that we model, blue polygons are areas of present-day continental lithosphere that are inferred to exist during the Neoproterozoic, but without having firm geological evidence or that have been affected by subsequent deformation (e.g. the distance between a present-day coastline and COB). Green polygons represent a schematic interpretation of congruent continental lithosphere, with intervening crust being subsequently deformed during future tectonic cycles. ADO = Aravalli-Delhi Orogen; Af, Afghanistan; Aus-Ant, Australia-Antarctica; Ca, ; DML-CLN, Dronning Maud Land-Coats Land Block; I, India; I-A, Indo-Antarctica; KNT (I/UM), Krygyz North Tianshan (Issendonian/Ulutau-Moyunkum); KMT (I/UM), Krygyz Middle Tianshan (Issendonian/Ulutau-Moyunkum); MIP, Malani Igneous Province; MO, Mirovoi Ocean; MzO, Mozambique Ocean; NC, North China; NeoMzO, Neomozambique Ocean; Qa, Qaidam; Qi, Qilian; Ru, Ruker; SC, South China; SM, Sahara Metacraton; T, Tarim; WAC, West African Craton; Y, Yangtze. 1233 1234 Early Tonian age (1000–900 Ma) magmatism and high-pressure metamorphism is preserved in the 1235 basements of the Qilian-Qaidam (Qi-Qa), Kunlun and Tianshan-Chu Yili terranes (Song et al., 2012; Tung 1236 et al., 2007; Wu et al., 2017). Importantly Song et al. (2012) identified an early Tonian event preserved in 1237 a high-pressure metamorphic belt in Qi-Qa. Here, a ~200 km linear belt of granitic , 1238 metamorphosed in the Palaeozoic, have crystallisation ages between ca. 1000 and 900 Ma (Song et al., 1239 2012). Zircons recovered from pelitic and psammitic gneisses from the same belt possess multiple 1240 generations of growth, as suggested through cathodoluminescence imaging, and return ages of the (first 1241 generation) to between ca. 940 and 900 Ma (Song et al., 2012). These are interpreted to represent a period 1242 of granulite facies metamorphism from a indicating that subduction was active during the 1243 Early Tonian (Song et al., 2012). Song et al. (2012) and others (e.g. Zhang et al., 2008) suggested a link 1244 between these two blocks and South China on the basis of similar-aged magmatism and metamorphism. 1245 However, as the Qi-Qa preserves a different Palaeozoic tectonic history to South China, as opposed to 1246 fragments of older lithosphere preserved in the Panxi-Hannan Belt of the Yangtze Craton, we suggest that 1247 a subduction zone was located outboard of Qi-Qa while a secondary, smaller ocean was closing between 1248 Qi-Qa and the accretionary orogen of the Panxi-Hannan Belt (Fig. 11a–c). Upon the suturing of Yangtze 1249 with Cathaysia (ca. 900 Ma, Fig. 11b), subduction relocated outboard of South China and began to close 1250 the ocean between Qa-Qi and South China. Similar to the Qi-Qa, the Kunlun terrane preserves scattered 1251 magmatic ages of S-type and protoliths of orthogneiss and amphibolites ranging between ca. 1000 1252 and 900 Ma (Chen et al., 2008; He et al., 2018; and Chen et al., 2006a; 2006b—both cited in Chen et al., 1253 2008; He et al., 2018). We interpret these rocks and ages as an extension of the same subduction zones 1254 outboard of Qi-Qa, and extend it further to the south where (again) similar-aged magmatism is also 1255 preserved in the North Tianshan and Chu-Yili (Degtyarev et al., 2017). 1256 1257 The Kazakh terranes (including Krygyz Tianshan and Chu-Yili, Fig. 15a) have poorly constrained 1258 Neoproterozoic histories, with only a handful of ages from outcropping magmatic rocks and sedimentary 1259 successions. We predominantly follow the summary of Degtyarev et al. (2017) in offering a possible 1260 tectonic interpretation of their Neoproterozoic geological history that is linked to the wider globe. 1261 Degtyarev et al. (2017) note that there are two broad categories of Precambrian terranes preserved in the 1262 Kyrgyz-Tianshan-Yuli area; the Issedonian and Ulutau-Moyunkum terranes. The Issedonian terranes, 1263 preserved in the northeast of the western Central Asian Orogenic Belt, include Chu-Yili and the Chinese 1264 Central Tianshan and are characterised by late Mesoproterozoic magmatism (e.g. Degtyarev et al., 2011), 1265 thick (> 1000 m) 1050–950 Ma quartzite-schist successions followed by ongoing magmatism from ca. 960– 1266 890 Ma (e.g. Degtyarev et al., 2008; Gao et al., 2015; Huang et al., 2015). Comparably, the Ulutau- 1267 Moyunkum terranes are preserved only in the west (in Krygyz) within Kyrgyz Middle Tianshan and Krygyz 1268 North Tianshan (e.g. Fig. 10b), and consist of a Palaeoproterozoic basement, with predominantly 1269 sedimentary Mesoproterozoic and early Neoproterozoic rocks (Degtyarev et al., 2017). Magmatism, 1270 between 840 and 760 Ma (Kröner et al., 2012) and granulite facies metamorphism from 800–760 Ma 1271 (Degtyarev et al., 2017; Tretyakov et al., 2016) are recorded only in the late Tonian. Both sets of terranes 1272 preserve distinct differences in their Mesoproterozoic histories, minor differences in the Early 1273 Neoproterozoic histories but similar histories from the mid-Neoproterozoic (ca. 700 Ma) onwards, 1274 suggesting proximity sometime during the late Tonian (800 to 700 Ma?) (Degtyarev et al., 2017). We 1275 interpret the Issedonian terranes to be the southernmost extent of the subduction zone spanning Qi-Qa and 1276 Eastern Kunlun, as the magmatism preserved in the Issedonian terranes has a continental arc signature 1277 (Huang et al., 2014) and is broadly coeval. Comparably, the Ulutau-Moyunkum terranes, which share 1278 Mesoproterozoic similarities to Tarim and record no magmatism in the early Neoproterozoic, are located 1279 on the opposite side (lower plate) of a closing ocean basin. This culminates with the collision of Tarim 1280 (with the Ulutau-Moyunkum terranes) and the combined India-South China continent at ca. 800–760 Ma 1281 along the Issendonian margin (Fig. 11d, e). The sparse data and age constraints from these terranes means 1282 much of their Neoproterozoic history is conjectural. Although the specific orientation and positioning of 1283 the terranes along the margin is speculative our interpretation is that it places the Kazakh terranes in a 1284 favourable position for their Palaeozoic evolution which is, comparably, much better constrained. 1285 1286 Two clusters of Neoproterozoic-aged palaeomagnetic data from Tarim make it difficult to elucidate a 1287 consistent position with other palaeomagnetic data from Rodinian constituents. Three poles from the Tonian 1288 and Cryogenian require a 90° rotation of Tarim in order to fit the younger pole of Wen et al. (2017), or a 1289 180° rotation to fit the cluster of three poles in the Ediacaran to Cambrian (see Merdith et al. (2017a) and 1290 Section 4.2 for a discussion). Within a self-consistent kinematic plate boundary framework, this motion is 1291 not permissible if Tarim is positioned either against north- (e.g. Zhang et al., 2012) or as 1292 an extension of the ‘Missing-Link’ model (Li et al., 2004 for the original proposal of the “Missing-Link” 1293 model; Wen et al., 2018, 2017). A plausible position where these palaeomagnetic criteria are met, along 1294 with satisfying key geological evidence, such as the 800–760 Ma Aksu Blueschist preserved on the northern 1295 margin of Tarim(C.-L. Zhang et al., 2013), is outboard of the India-South China accretionary belt, where it 1296 acts as the final piece of continental lithosphere accreted to the margin. In our model, we suggest the 1297 metamorphism recorded by the Aksu Blueschist marks the accretion of Tarim to Chu-Yili and the Tianshan 1298 (see Xia et al. (2017) for the most recent discussion, but see also data and discussion from Zhang et al. 1299 (2012, 2009). To accommodate the change in position suggested by the palaeomagnetic data, we introduce 1300 a ~120° rotation of Tarim and Southern Tianshan away from this margin, such that Tarim’s southern margin 1301 collides with the outboard margin of Alxa-Qaidam-Qilian, so its northern margin faces an open ocean basin, 1302 allowing it to rift northward facing as Gondwana forms towards its more well constrained Palaeozoic 1303 position (Fig. 16). 1304 1305 To support this interpretation of Tarim’s evolution we present the following geological observations in 1306 support of this model. Firstly, there is an absence of extensive magmatism on either the northern or southern 1307 margin of the Tarim craton between 1000 and 850 Ma, which makes it difficult to include as a part of the 1308 upper-plate circum-Rodinian subduction girdle (Cawood et al., 2016). Previous studies, including MER17, 1309 place Tarim on the margin of Rodinia typically also include the Chu Yili and Tianshan crust attached in a 1310 quasi-present-day configuration to the northern margin of Tarim. In such cases their record of Tonian 1311 magmatism supports the interpretation that they formed part of the circum-Rodinian subduction girdle (Ge 1312 et al., 2014) however, this location is inconsistent with available palaeomagnetic data (e.g. (Wen et al., 1313 2018). Secondly, rift related granitoids preserved in the Southern Tianshan (Degtyarev et al., 2017; Gao et 1314 al., 2015) are here interpreted as evidence of the re-adjustment of Tarim between 730 and 680 Ma to account 1315 for the change in palaeolatitude and orientation inferred from palaeomagnetic data (Fig. 11e). In addition, 1316 extensive rifting in the southwest of Tarim (Wang et al., 2015) is interpreted to reflect the rearrangement 1317 of subduction after Tarim/India-South China amalgamation. This motion is similar to the adjustment of 1318 Baltica relative to Laurentia in the latest Mesoproterozoic proposed by Cawood et al. (2010) to account for 1319 the Valhalla . Xiao et al. (2010) summarise the geology and of rocks found in the 1320 Beishan area of China which record a protracted and complex history of multiple arc development and 1321 accretion through the late Neoproterozoic and Early Palaeozoic. Rocks in the Beishan area (Alxa, Fig. 15, 1322 16) range from low-grade sedimentary metamorphic assemblages to and complexes and 1323 intrusive granitic bodies that have late Neoproterozoic–Cambrian ages (Xiao et al., 2010). These rocks, 1324 inferred to represent an active subduction zone and accretionary complex, are unconformably overlain by 1325 Cambrian–Ordovician sediments. The earliest record of metamorphism in the area are from a series of 1326 SHRIMP U-Pb ages taken from the metamorphic rims of zircons of an eclogite unit at ca. 830–800 Ma 1327 (Yang et al., 2006). 1328 1329 With respect to the palaeomagnetic issues of Tarim we raised earlier (Section 4.2), we find that our 1330 conceptual model of this rotation of Tarim (e.g. Fig. 16) can fit either set of palaeomagnetic data equally as 1331 well, with the key factor being the time of subduction in the Beishan area. Under the cluster of three poles, 1332 the motion of Tarim occurs more quickly and peak subduction (possibly resulting in a collision?) would be 1333 earlier (ca. 650 Ma), while to fit the pole of Wen et al. (2017) it would occur later at between 600 and 1334 550 Ma. Based on the review of Xiao et al. (2010), we infer that the geological data support the later 1335 interpretation more strongly however, given the novelty of this scenario and the absence of identifiable 1336 piercing points, it could be revised in the future to fit the alternative scenario. 1337 1338 Late Neoproterozoic–Early Cambrian rifting events are inferred to have occurred within all the terranes 1339 (Kunlun, Qa-Qi, Chu Yili and Tianshan) that we have placed on this Indian-South Chinese accretionary 1340 margin, however, the high degree of reworking and suturing of crust from the terranes, coupled with the 1341 small size of these terranes makes it difficult to pin down precise rift times. We stress that our interpretation 1342 here is preliminary, especially when compared to specialised reviews of the tectonics of this area (Kroner 1343 et al., 2007; Wilhem et al., 2012; Windley et al., 2007; Xiao et al., 2013; Yakubchuk, 2017). We reiterate 1344 that our intention here is to provide a possible framework that connects and contextualises these terranes 1345 within a consistent kinematic and tectonic evolution between the Neoproterozoic and Palaeozoic (Fig. 16), 1346 which can be more tightly refined in the future. Ordovician–Silurian sutures between Qa-Qi-Kunlun and 1347 surrounding cratons preserve late Neoproterozoic–Cambrian ophiolites (Jian et al., 2014; Shi et al., 2018; 1348 Song et al., 2013, 2009) thus necessitating the existence of ocean basins, but there are few dates of ocean 1349 basin initiation. Xu et al. (2015) date Qilian-Qaidam continental rift basalts to 600–580 Ma, constraining 1350 ocean basin formation to the latest Ediacaran–earliest Cambrian, with the oldest ophiolite (the Yushigou 1351 Ophiolite) preserved in these terranes dated to 550 Ma (Shi et al., 2004). Evidence of rifting is more sparse 1352 in Kunlun, however, similar stratigraphy, ages and geochemistry of metavolcanic deposits between Kunlun 1353 and Qaidam-Qilian (Yuan et al., 2004) suggest coeval rifting is reasonable, though not definite. We follow 1354 some recent work (Peng et al., 2019; Zhao et al., 2018) in having separation of Tarim and these terranes in 1355 the late Ediacaran (550 Ma here, as a response to the closure of the Mozambique Ocean between India and 1356 Congo). Our model closely resembles the schematic framework outlined by Qiantao et al. (2001), while 1357 also following Domeier (2018) in maintaining a very close affinity between Kunlun, Q-Q and Alxa for the 1358 Early Palaeozoic, such that they conceptually form a single elongated terrane that rifts off Gondwana at ca. 1359 550 Ma and collides along the southern margin of Tarim by 440 Ma. 1360 1361 Comparably, rifting on the northern margin of Tarim is easier to constrain. Thick, late Neoproterozoic 1362 sedimentary sequences capped with carbonate and an on the Ediacaran–Early Cambrian 1363 imply a prolonged rift phase with breakup at ca. 550–540 Ma (Zhao et al., 2014; Zhu et al., 2017). Similarly, 1364 sedimentary assemblages preserved in the South Tianshan orogen are dated from 540–520 Ma (Alexeiev et 1365 al., 2020; Safonova et al., 2016), suggesting that ocean basin formation between Tarim and Krygyz Middle 1366 Tianshan begins in the Early–Middle Cambrian. Further north, the relationship between the Kryguz North 1367 Tianshan and Chu-Yili is also reasonably well established (e.g. Windley et al., 2007; Xiao et al., 2013). 1368 Late Ediacaran–Early Cambrian magmatism preserved in Chu-Yili and North Tianshan (Alexeiev et al., 1369 2011; Degtyarev et al., 2017; Kroner et al., 2007) is interpreted to represent development of multiple 1370 contemporaneous arcs (e.g. Alexeiev et al., 2020). The Ordovician-aged sutures between Chu-Yili and 1371 North and Central Tianshan (Windley et al., 2007) are defined by ophiolitic slithers, implying an ocean 1372 basin (or a back-arc) existed between these terranes. We infer that in the late Neoproterozoic to Early 1373 Cambrian, the Kazakh and Tianshan terranes that accreted outboard of India-South China were fragmented 1374 and rifted off this margin forming a collage (not dissimilar to modern , or the NE Pacific in 1375 the Mesozoic, e.g. Sigloch and Mihalynuk (2013) that eventually re-assembled in the Ordovician. We 1376 model the time of fragmentation at 550 Ma, because this is the time of collision between India and Congo 1377 along the East African Orogen, so we infer that subduction relocated outboard of the northern margin. 1378 1379 Palaeomagnetic data from South China do not permit a fixed fit between South China and Gondwana for 1380 the early Palaeozoic. The data do, however, permit a close spatial relationship between them (<1000 km). 1381 Palaeomagnetic data from South China suggest it moved from mid to lower latitudes between the Cambrian 1382 and Devonian (Domeier, 2018; Han et al., 2015). Furthermore, shallow marine faunal data and detrital 1383 zircon arrays suggest that between the Cambrian and Devonian, South China shifted from an Indian- 1384 Himalayan-Iran affinity (Burrett et al., 1990) to Sibumasu-Australian affinity (Cocks and Fortey, 1997; 1385 Metcalfe, 2013, 2011), broadly consistent of the positions necessary to fit the palaeomagnetic data. Rift 1386 sequences in South China and in northern India (Himalayan terranes) are similar, but diverge strongly after 1387 the early Cambrian (Jiang et al., 2003), providing some kinematic support for invoking for relative motion 1388 between South China and Gondwana. The Sanya Block of Hainan (Fig. 15f) is linked through detrital 1389 zircon provenance and middle Cambrian to western Australia and Antarctica rather than South 1390 China in the Neoproterozoic (Cawood et al., 2018b; Xu et al., 2014). The presence of Early Ordovician 1391 trilobites in the Sanya block also common to South China and Australia (Torsvik and Cocks, 2009) support 1392 a close relationship between the three domains by this time (Cawood et al., 2018b). A late Cambrian– 1393 Ordovician (520–450 Ma) metabasaltic arc assemblage in Hainan Island (Xu et al., 2008, 2007) is 1394 interpreted to be the northerly extension of the Kungaan Orogen that sutured Australia-Antarctica and India 1395 (e.g. Xu et al., 2014). We introduce divergent motion between Gondwana and South China at 550 Ma, 1396 coinciding with the rift-to-drift sequences of Jiang et al. (2003). This divergent motion moves South China 1397 from a position outboard of Northern India at 550 Ma to one that is slightly further east and adjacent to 1398 western Australian at 500 Ma, accounting for the arc assemblage in Hainan as well as similar faunal 1399 patterns.

Figure 16 Schematic of our model for India-South China at key time steps, showing accretion of the Yangtze Craton and numerous smaller terranes and blocks to a large, Tonian subduction zone outboard of the north- western India and the northern margin of South China and their subsequent fragmentation and rifting off during the Ediacaran–Early Cambrian. (a) early Tonian; (b) late Tonian; (c) Cryogenian–Ediacaran, note if the model were adopted to fit the cluster of three palaeomagnetic poles from Tarim then collision would be at ca. 650 Ma, and; (d) Ediacaran–Cambrian. Af, Afghanistan; Al-Qi-Qa, Alxa, Qilian, Qaidam; ANS, Arabian Nubian Shield; CY, Chu Yili; DML-CLB, Dronning Maud Land-Coats Land Block; EAO, East African Orogen; I, India; I-A, Indo-Antarctica; KMT (I/UM), Krygyz Middle Tianshan (Issendonian/Ulutau-Moyunkum); KNT (I/UM), Krygyz North Tianshan (Issendonian/Ulutau- Moyunkum); SC, South China; T, Tarim. 1400 1401 5.4.3 ANS-Azania-TOAST 1402 We suggest that, to a first order, the central Arabian-Nubian Shield (ANS) accreted on the kernel of Azania 1403 and formed a semi-continuous archipelago outboard of the eastern margins of the Congo Craton and Sahara 1404 Metacraton (SM) (Fig. 17; Collins and Pisarevsky, 2005; Merdith et al., 2017a). Geological details and a 1405 regional plate model of the accretion of the ANS and Azania are adopted from Collins et al. (in revision), 1406 Blades et al. (2019a) and Johnson et al. (2011), though here we extend Azania to the south by attaching 1407 portions of the Tonian Aged Ocean Arc Super Terrane (TOAST—Jacobs et al. (2017, 2015)). The similarity 1408 in ages, petrology of rocks and δ18O from zircons between the Dabolava Suite in Madagascar (Archibald et 1409 al., 2018) and TOAST (Jacobs et al., 2017, 2015; Wang et al., 2020) suggest a similar tectonic environment. 1410 As the southern tip of Azania is reconstructed to be adjacent to the location of the TOAST terrane in 1411 Gondwana, at the nexus between the East African Orogen and the Pinjarra/Kuunga Orogen, there is also a 1412 strong palaeogeographic argument for attaching TOAST to Azania, as their Rodinian reconstructed position 1413 requires no alteration for their position in Gondwana (Fig. 17b). 1414

Figure 17 Amalgamation of the Arabian-Nubian Shield (ANS) after Blades et al. (2019b, 2015), Armistead et al. (2019) and Collins et al. (submitted) and the incorporation of the ANS, Azania and TOAST into Gondwana along the East African Orogen at: (a) 800 Ma and (b) 520 Ma. Tan polygons are areas of continental lithosphere in the Neoproterozoic that we model, blue polygons are areas of present-day continental lithosphere that are inferred to exist during the Neoproterozoic, but without having firm geological evidence or that have been affected by subsequent deformation (e.g. the distance between a present-day coastline and COB). Green polygons represent a schematic interpretation of congruent continental lithosphere, with intervening crust being subsequently deformed during future tectonic cycles. Thick black lines follow the suture of the East African Orogen and Kuunga Orogen. ANS, Arabian-Nubian Shield; EAO, East African Orogen; SF, Sao Francisco; SM, Sahara Metacraton. 1415 1416 5.4.4 Hoggar, Borborema, Avalonia and Ganderia 1417 The Hoggar Block is preserved between the Sahara Metacraton (SM) and the West African Craton (WAC) 1418 in northwest Africa and records a long Neoproterozoic history of accretion of island arcs and continental 1419 ribbons. The model incorporated here is based on fieldwork by Caby et al. (1989), Black et al. (1994) and 1420 Liégeois et al. (1994) and involves three main constituents of present-day Hoggar (from west to east): 1421 IOGU-IGU, LATEA and the Aïr Block (Fig. 18a, b). 1422 1423 The broad tectonic framework of the Hoggar block is an accretionary margin consisting of at least 23 1424 individual terranes that were slowly compressed between two large tectonic units—the WAC and the SM— 1425 as Gondwana amalgamated. The Aïr block, preserved in the east, is an amalgamation of three closely related 1426 terranes: the Aouzegueur, Barghot and Assodé-Issalane terranes. The first two terranes accreted onto the 1427 margin of the SM by 650 Ma, with the Aouzegueur terrane preserving a tonalite-trondhjemite- 1428 (TTG) suite dated at ca. 730 Ma and the Barghot terrane recording calc-alkaline granitoids from ca. 730 to 1429 660 Ma, with a post nappe pluton preserving a U-Pb zircon age of 664±8 Ma, interpreted to provide a 1430 minimum age for deformation (Liégeois et al., 1994). Both these terranes were metamorphosed to 1431 greenschist or amphibolite facies and were cut by east-verging thrusts. In contrast, the Assodé-Issalane 1432 terranes exhibit younger magmatism (ca. 640–580 Ma) and amphibolite-facies metamorphism. They are 1433 thrust east over the Barghot terrane. Both Black et al. (1994) and Liégeois et al. (1994) suggest that the 1434 Aouzegueur terrane collided first with the SM, followed by the Barghot terrane, which is positioned slightly 1435 further south than the former, through an east-dipping subduction zone underneath the two terranes (Fig. 1436 18). Following collision, the Assodé-Issalane terrane (which until this time we position slightly west of the 1437 former terranes) was thrust above of the Barghot terrane in response to the closure of the ocean between 1438 the WAC and the SM. Our reconstruction implies that this motion was predominantly transpressive, along 1439 the dextral Raghane shear zone with plutons dating from ca. 630 to 580 Ma (Liégeois et al., 1994). 1440 1441 Further west from the Aïr block in central Hoggar, the LATEA terranes (Laouni, Azrou-n-Fad, Tefedest 1442 and Egéré-Aleksod) all consist of Archaean to Palaeoproterozoic basement, but preserve no 1443 Mesoproterozoic or early Neoproterozoic rocks. LATEA was a passive cratonic unit for most of the 1444 Neoproterozoic and acted as a nucleus for the accretion of juvenile terranes. The earliest Neoproterozoic 1445 activity is the accretion of the ca. 900 Ma juvenile Iskel to the western margin of LATEA, with 1446 subduction inferred to be west dipping away from LATEA (Liégeois et al., 2003). The protolith of an 1447 eclogitic unit, currently preserved along the shear zone delineating the Iskel arc and LATEA, is dated to ca. 1448 870 to 850 Ma by U-Pb dating from zircons extracted from syn-to-late kinematic plutons (Caby et al., 1449 1982). The combined In-Ouzzal and Iforas granulite units (IOGU/UGI), which are Palaeoproterozoic 1450 continental ribbons, preserve few Tonian rocks. From ca. 700 to 640 Ma magmatism is recorded throughout 1451 the entire region, suggesting that subduction occurred along both margins of the terranes (Caby, 2003) (Fig. 1452 18d, e). At 630 Ma, collision between the IOGU/UGI terranes and the LATEA block occurred, forming the 1453 combined present-day central-western Hoggar region. 1454 1455 The final tectonic events of this area involve a two-step amalgamation process of Western and Central 1456 Hoggar (i.e. IOGU/UGI and LATEA) to the Aïr Shield and SM, and the collision between this landmass 1457 (Hoggar and the SM) and the WAC, where collisional deformation is preserved in the Pharusian and 1458 Dahoymede belts (Merdith et al., 2017a) (Fig. 18f, g). Here, subduction is inferred to have occurred away 1459 from the WAC (i.e. underneath Hoggar) due to the absence of magmatic rocks preserved on the WAC. 1460 Continual dextral deformation is preserved throughout Hoggar until ca. 530 Ma, suggesting that there was 1461 still relative motion until the Cambrian (Liégeois et al., 2003; Paquette et al., 1998). The major regime of 1462 the Hoggar block between the Ediacaran and early Cambrian was transpressive, resulting in extensive 1463 faulting and upwelling of the ; causing partial melting of lower Archaean crust in some areas 1464 (Hadj-Kaddour et al., 1998). We interpret the final amalgamation of Hoggar to occur at ca. 580 Ma, as this 1465 is the age of the syntectonic, deformed plutons found among the shear zones that bind the SM and the WAC. 1466 Here, a suite of ages include: Rb-Sr whole ages from dykes affected by the transpressive event yield 1467 an age of 592±6 Ma (Hadj-Kaddour et al., 1998); a 583±7 Ma U-Pb age on zircons extracted from the syn- 1468 to-late tectonic, elongated Imezzarene pluton (Lapique et al., 1986); a 594±4 Ma and 593±17 Ma U-Pb age 1469 of zircon extracted from the Ohergehem and Adaf plutons, respectively (Henry et al., 2009). Fezaa et al. 1470 (2010) identified younger (ca. 575–555 Ma) deformation in the Murzoq area of Hoggar, however, they 1471 suggested that it was unrelated to the main convergence between WAC and SM. 1472

Figure 18 Amalgamation of Hoggar after Caby et al. (1989), Black et al. (1994) and Liégeois et al. (1994), Borborema and Peri-Gondwanan terranes after Nance et al. (2008) at key time slices. (a) map of key terranes in a reconstructed Gondwana (dark blue terranes are oceanic terranes); (b) map of reconstructed Peri-Gondwanan terranes at 650 Ma; (c) 1000 Ma; (d) 900 Ma; (e) 860 Ma; (f) 700 Ma; (g) 640 Ma; (h) 580 Ma; (i) 550 Ma and (j) 520 Ma. Tan polygons are areas of continental lithosphere in the Neoproterozoic that we model, blue polygons are areas of present-day continental lithosphere that are inferred to exist during the Neoproterozoic, but without having firm geological evidence or that have been affected by subsequent deformation (e.g. the distance between a present-day coastline and COB). Green polygons represent a schematic interpretation of congruent continental lithosphere, with intervening crust being subsequently deformed during future tectonic cycles A, Aouzegueur; A-I, Assodé-Issalane; Am, Amazonia; ANS, Arabian-Nubian Shield; B, Baltica; BP, Borborema Province; C, Congo; C-B, Central Borborema; IGU, Iforas granulite unit; IOGU, In Ouzzal granulite unit; IO, Iapetus Ocean; LATEA, Laouni, Azrou-n-Fad, Tefedest and Egéré-Aleksod terranes; MO, Mirovoi Ocean; Moz. O, Mozambique ocean; N-B, Niger-Benin Block; P, Paranapanema; PG, Peri-Gondwanan terranes; S-B, Southern Borborema; SF, Sao Francisco; SM, Sahara Metacraton; WAC, West African Craton. 1473 1474 The Borborema block sits between Congo-SF and the Nigeria-Benin Shield in a reconstructed Gondwana 1475 (Fig. 18a). This small block consists of Archaean–Proterozoic basement that was strongly reworked and 1476 deformed during the Gondwana amalgamation events between Africa and Amazonia (dos Santos et al., 1477 2010). Magmatism in the Transversal Domain of Central Borborema is thought to represent a local, early 1478 Tonian orogeny called the Cariri Velhos Orogeny (da Silva Filho et al., 2002; dos Santos et al., 2010). Late 1479 Stenian rift deposits are preserved in the Cariri Velhos belt, suggesting that the Pernambuco-Alagoas 1480 domain (PEAL) (the crystalline basement of Southern Borborema) originally rifted from this margin 1481 (Guimarães et al., 2012), before the ocean inverted and closed. Our model places the Northern and 1482 Transversal Domains of Borborema fixed to the Nigeria-Benin-SM blocks, while the PEAL closed the 1483 small Stenian aged relict ocean basin as it collided with the northern and central Borborema provinces by 1484 ca. 920 Ma (Caxito et al., 2014a), forming the Cariri Velhos Orogen. An ocean basin remained on the 1485 southern margin of PEAL until the Ediacaran, as the final collision between Borborema and SF did not 1486 occur until this time forming the Sergipano belt, which preserves relict oceanic crust (e.g. Caxito et al., 1487 2014b; Ganade de Araujo et al., 2014) (e.g. Fig. 9). To accommodate the Tonian closure, while maintaining 1488 an open ocean basin to the south of PEAL, we follow Caxito et al. (2014a, 2016) who suggested that the 1489 synchronous rifting and formation preserved in SF (Pedrosa-Soares et al., 2001), which in our 1490 model is reconstructed to be adjacent to PEAL, are relicts of the divergent motion necessary to achieve this. 1491 1492 The Avalonian terranes, currently preserved in the east coast of modern day North American and western 1493 Europe, have a well-documented Neoproterozoic history (e.g. Murphy and Nance, 1989). The Avalonian 1494 terrane is interpreted to be underlain by ca. 1.0 Ga juvenile basement on the basis of 1.3–0.8 Ga Sm-Nd 1495 depleted mantle model ages in in younger Neoproterozoic rocks (Murphy et al., 2000; Thorogood, 1990). 1496 These younger rocks consist of magmatic gneiss and plutonic complexes, along with tuffs, pelitic schists 1497 and quartzites. U-Pb ages of the complexes and tuffs range from ca. 750–650 Ma (Bevier et al., 1993; Doig 1498 et al., 1993; Keppie and Dostal, 1998; Krogh et al., 1988; O’Brien et al., 2001) and detrital zircons from 1499 the (meta-)sedimentary rocks suggest they were sourced from a juvenile arc (metapelites, Murphy, 2002) 1500 and a cratonic source (quartzite), typically inferred to be Amazonia or the WAC (Nance et al., 2008). This 1501 magmatism is followed by amphibolite–granulite metamorphism from 660–630 Ma (Keppie et al., 1998; 1502 Strachan et al., 2007). Younger magmatism (ca. 640–550 Ma) is more voluminous and includes abundant 1503 arc derived volcanic, plutonic complexes and coeval volcanic-sedimentary successions (Bevier et al., 1993; 1504 Compston et al., 2002; Doig et al., 1993; Nance et al., 2008; O’Brien et al., 2001; White et al., 2020). 1505 Subduction does not continue into the Cambrian, instead a clastic platform and transition to rift environment 1506 begins to form, culminating in the opening of the Rheic Ocean (Domeier, 2016; Nance et al., 2008; Nance 1507 and Linnemann, 2008). 1508 1509 To a first order, Ganderia and Carolina, both preserved in North America, record a similar Neoproterozoic 1510 history to Avalonia, differing predominantly in that key metamorphic and magmatic events are ca. 30– 1511 40 Ma younger than in Avalonia (Hibbard et al., 2007; Nance et al., 2008; van Staal et al., 2012). Depleted 1512 mantle model ages from Sm-Nd isotopes also hint at the presence of ca. 1.2–0.8 Ga juvenile crust in 1513 Carolina (Hibbard et al., 2007). However, the principal period of magmatism in both Ganderia and Carolina 1514 is preserved from 650 to 580 Ma and is inferred to have occurred in an ocean-arc environment (Hibbard et 1515 al., 2007). Metamorphism, up to eclogite facies, occurs at the end of this period and continues into the 1516 earliest Cambrian (580–540 Ma) (Barker et al., 1998; Shervais et al., 2003). Younger magmatism (to ca. 1517 520 Ma, White et al., 2002) linked to the rifting and opening of the Rheic Ocean, is only preserved in 1518 Ganderia (Hibbard et al., 2007). Finally, the Suwannee terrane of Florida is linked tectonically to both the 1519 West African Craton and Amazonia throughout the Neoproterozoic (Dallmeyer, 1989), lacking the 1520 Neoproterozoic arc development preserved in Avalonia, Ganderia and Carolina. Instead, 550 Ma calc- 1521 alkaline volcanic rocks are inferred to represent the remnants of a continental arc (Heatherington et al., 1522 1996). 1523 1524 We follow the model of Nance et al. (2008) and Murphy et al. (2004) for the Neoproterozoic evolution of 1525 these terranes. An early Tonian (1–0.8 Ga) oceanic arc outboard of Baltica-Amazonia-WAC dipping under 1526 the Rodinian plate, formed the earliest portions of crust preserved in these terranes (Fig. 18c–e). The relative 1527 positioning of the terranes follows that of DOM16, with East Avalonia most easterly (Fig. 18b), then West 1528 Avalonia, and Ganderia and Carolina furthest west, with Ganderia sitting oceanward of Carolina. This 1529 arrangement follows the same logic outlined in Section 5.2.2, as by maintaining this relative positioning 1530 we avoid having to laterally re-organise the terranes during the Palaeozoic. This is slightly different from 1531 the positioning in Nance et al. (2008), who model West Avalonia more easterly (relative to a fixed WAC) 1532 than East Avalonia and invoke wrench-tectonics to laterally translate the terranes. Nonetheless, at ca. 750 1533 Ma this subduction polarity reversed and the adjacent ocean basin between WAC-Baltica began to subduct 1534 underneath Avalonia. We model Ganderia and Carolina slightly further away from the active subduction 1535 front–behind Western Avalonia, to account for their lack of Cryogenian magmatism. This arc front collided 1536 with WAC at ca. 650 Ma when the subduction ceased and reset outboard of the now amalgamated 1537 Avalonian-WAC continent as a continental arc (Fig. 18f, g). This subduction continued until the Cambrian 1538 when the area transitioned into a rift environment when the Rheic Ocean opened. 1539 1540 This last phase our model is preliminary and needs further development, though we hope that it provides a 1541 framework that can assist with testing alternative scenarios. By fitting the latitude of the ca. 520–500 Ma 1542 poles of Baltica (Section 5.2.1), subduction must consume most of the relic ocean basin immediately north 1543 of the Avalonian margin of Gondwana between 550 and 520 Ma. The magmatism in Ganderia easily 1544 accounts for this, however Baltica needs to be further east at 520 Ma, otherwise it must undergo 4000 km 1545 of dextral motion (relative to Gondwana) to allow for the initial stages of the Rheic Ocean opening at ca 1546 500 Ma (e.g. Domeier, 2016; von Raumer and Stampfli, 2008), which we suggest is not a reasonable 1547 scenario. As such, our model places a subduction zone slightly outboard of Avalonia, but acknowledge that 1548 this is a simplification that needs further refinement. 1549 1550 6 Plate Model 1551 1552 Having discussed the motions of the evolution of continental configurations in previous sections, here we 1553 discuss the more speculative elements of the reconstruction—the oceanic plates and plate boundaries. For 1554 further in-depth discussion of the continental portions of the model, in particular the major Gondwana 1555 forming sutures and evolution of the post-Cambrian world, we point readers to the studies that produced 1556 the base models used here (MER17, DOM16/18, YOU19, DT14). We also provide in our supplementary 1557 material the associated plate model files (SM2), as well we a tectonic summary of seafloor production and 1558 consumption rates, mid-ocean ridge length and subduction zone length (SM3). 1559 1560 6.1 Synthetic ocean plates 1561 1562 The construction of synthetic ocean plates is required to maintain tectonic congruency (Section 2.7) but, 1563 with few exceptions (e.g. ophiolites), there is no direct evidence of the configuration or tectonic parameters 1564 (e.g. spreading rate, asymmetry) of oceanic crust for the pre-Mesozoic due to the constant subduction of 1565 oceanic lithosphere. However, we know that oceanic crust typical of present-day (i.e. MORB) did exist in 1566 the Palaeozoic and Neoproterozoic, as evidenced by ophiolitic remains preserved in orogens (Furnes et al., 1567 2014). We therefore use one key assumption when constructing oceanic plates: we assume that the 1568 production (rate of motion, orthogonal spreading etc.) and subduction of oceanic crust in the 1569 Neoproterozoic was fundamentally similar to the Cenozoic. We note that this may not be a valid assumption 1570 for the early Neoproterozoic, since abundant ophiolite preservation only occurs after Rodinia breakup 1571 (Stern and Miller, 2018) and pre-1 Ga ophiolites suggest thicker oceanic crust (Moores, 2002) which, along 1572 with secular changes in Earth’s heat loss (Brown et al., 2020b) could have an influence on spreading and 1573 subduction dynamics. Nonetheless, we maintain that if available palaeomagnetic and geological data can 1574 be reconciled within a uniformitarian framework of oceanic crust production and destruction, then our 1575 model becomes a useful reference model for future models that explore alternative hypotheses. 1576 Measurements of seafloor production, crustal consumption, ridge length and subduction zone length are 1577 provided in the supplementary material. 1578 1579 6.1.1 Early Tonian until Rodinia breakup (1000-750 Ma) 1580 1581 In our model, three prominent ocean basins existed in the early Tonian: the Mirovoi Ocean (McMenamin 1582 et al., 1990; Meert and Lieberman, 2008), the Mawson Sea (Meert, 2003; Meert and Lieberman, 2008) and 1583 the Mozambique Ocean (Fig. 8) (Collins et al., 2003; Collins and Pisarevsky, 2005). These ocean basins 1584 have been defined previously in the same context as they appear in our model, however, given differences 1585 between our model and the original publications, the geographical boundaries of each ocean are slightly 1586 different. 1587 1588 We define the Mirovoi Ocean as the large ocean bordering Rodinia in its west and India-South China and 1589 the Sahara Metacraton in the north east and south east respectively. The Mirovoi is the largest and most 1590 prominent ocean basin for the Neoproterozoic in our model, existing until ca. 520 Ma with the opening of 1591 the Proto-Tethyan Ocean and Ran Sea (Fig. 9, Hartz and Torsvik, 2002). It is (conceptually) equivalent to 1592 the external Panthalassic and Pacific oceans of the Phanerozoic, as it consists almost entirely of oceanic 1593 lithosphere and is ringed by subduction for the majority of its existence. At 1000 Ma, we model a triple 1594 junction spreading ridge located roughly in the centre of the ocean basin. The triple junction provides three 1595 directions of spreading to account for convergence in three areas: (i) the closure of the ocean basin 1596 separating India-South China from Tarim, Qaidam-Qilian, Lut, Afghanistan, Kunlun and Tarim (this study); 1597 (ii) the Taimyr subduction zone outboard of northern Siberia (Metelkin et al., 2012; Vernikovsky et al., 1598 2004) and the Valhalla Orogen outboard of Greenland (Cawood et al., 2010), and; (iii) the Proto-Avalonian- 1599 Cadomian subduction zone outboard of Baltica and Amazonia (Murphy et al., 2000) that extends 1600 northwards to Sao Francisco and further north where it is preserved in the Iskel Island Arc of Hoggar 1601 (Liégeois et al., 2003). 1602 1603 The spreading arms of the triple junction span north, east-southeast and southwest. The northern arm 1604 separates Siberia and the India-China accretionary zone and extends partway into the Mawson Sea. The 1605 east-southeastern arm extends towards India, where we connect it via a to the mid-ocean 1606 ridge in the Mozambique Ocean. The southwestern arm intersects the Proto-Avalonian subduction zone. 1607 We model this configuration (triple junction) as being stable through the early Tonian until ca. 870–850 1608 Ma, where a plate-reorganisation event occurs. We link this organisation to a change in kinematics of 1609 Rodinia, suggested through palaeomagnetic data. At this time, palaeomagnetic data from the Baltica 1610 (Walderhaug et al., 2007) suggest that Rodinia had drifted to southerly latitudes, before returning to 1611 equatorial latitudes by ca. 750 Ma (Eyster et al., 2019). MER17 modelled significant relative dextral motion 1612 between Congo-SF-Azania and Rodinia during this time (870–750 Ma) as well, as suggested by 1613 palaeomagnetic data (Evans et al., 2016). Geological data from Congo-Azania also supports the rotation, 1614 with onset of sedimentation interpreted to be a rift event in the Damara region (Armstrong et al., 2005; 1615 McGee et al., 2012) and the onset of a massive subduction system outboard of Azania (Archibald et al., 1616 2017; Handke et al., 1999). During this transition (850–800 Ma), we model the Mirovoi Ocean as a single 1617 spreading ridge, orientated sub-parallel to the north-facing arm in the original triple junction, extending 1618 northwards toward Siberia and intersecting an oceanic arc outboard of Baltica. 1619 1620 For the Tonian, until 750 Ma, the South China-India continent moved to polar latitudes on the north-eastern 1621 side of the Mirovoi Ocean basin. Comparably, Siberia started dextral motion relative to Rodinia, from a 1622 position near Australia-North China to one near Greenland (Pisarevsky et al., 2013). To account for this 1623 motion, alongside ongoing subduction in the Taimyr region of Siberia, we have extended the spreading 1624 ridge from the Mozambique Ocean into the Mirovoi Ocean (running E-W) and have a northern arm 1625 accounting for divergence between Tarim and the Taimyr subduction zone. This interpretation necessitates 1626 the presence of a triple junction, with a third ridge arm intersecting the subduction zone outboard of Baltica, 1627 similar to the configuration at the start of the Tonian. 1628 1629 The Mawson Sea is defined as the ocean basin between Australia and India-South China that closed with 1630 the amalgamation of Gondwana (Meert, 2003). In the early Tonian, this basin is large in our model, 1631 necessitated by MER17s removal of India-South China from Rodinia. The large size is because relative 1632 longitude prevents India-South China (at this point in time travelling north from the equator) from being 1633 any closer to Australia, as Azania occupied the same latitude as India and lay between India and Australia. 1634 We model a single spreading ridge in the centre of the ocean basin, accounting for the subduction on the 1635 India-South China margin as suggested in the Eastern Ghats of India and accretion of the Ruker Terrane to 1636 Indo-Antarctica (e.g. Corvino et al., 2008; Liu et al., 2017). In South China, this subduction is more sparse, 1637 but recent work has suggested that part of present-day Vietnam is associated with the southwestern Yangtze 1638 craton (Minh et al., 2020) and could record the late Tonian portion of a subduction system and be the focus 1639 of future work. At 930 Ma the spreading direction of the ridge changes, to compensate for the southerly 1640 drift of Rodinia. The change in spreading direction coincides with the docking of North China-Australia- 1641 Antarctica with Laurentia as the final amalgamation event of Rodinia (Mulder et al., 2018b). At this time 1642 we model subduction initiating outboard of Australia (against Lhasa, (e.g. Guynn et al., 2006) and into the 1643 ocean outboard of Antarctica further south. Preserved evidence of subduction here is sparse due to ice cover 1644 in Antarctica. However, this area of the ocean (just outboard of the western margin of the Mawson craton) 1645 is positioned at the centre of Antarctica in the nexus between the TOAST terrane and Mawson craton, so it 1646 could be possible some arcs are preserved in Antarctica. Further south, this subduction zone transitions into 1647 a transform boundary and separates relative motion between Azania-Congo and Rodinia between 930 Ma 1648 and 850 Ma. We model continual spreading in the Mawson Sea until 850 Ma, at which time the ocean basin 1649 begins to close and we do not model an active ridge. By 750 Ma the Mawson Sea is extremely narrow, with 1650 only 2500–3500 km of ocean basin separating Australia and India. A narrow ocean basin is supported 1651 geologically by the strong evidence suggesting that a large sinistral shear zone was present outboard of 1652 western Australia and Antarctica during the Cryogenian and Ediacaran, suggesting that there was close 1653 proximity without collision between Australia-Antarctica and another continent (Collins, 2003; Fitzsimons, 1654 2003; Halpin et al., 2017; Merdith et al., 2017b; Powell and Pisarevsky, 2002). 1655 1656 The Mozambique Ocean is described as the ocean that closed with the collision of India and Congo along 1657 the East African Orogen reacted to Gondwana amalgamation during the Cryogenian and Ediacaran (e.g. 1658 Collins and Pisarevsky, 2005). For the sake of continuity, we refer to this ocean as the Mozambique in the 1659 Tonian as well. A small ocean (in our model, roughly equivalent in size to the between 1660 Australia and New Zealand), termed ‘Neomozambique Ocean’ also closed with the formation of the East 1661 African Orogen, however this ocean was located between ANS-Azania-TOAST and Congo (Fig. 8). 1662 1663 Geographically, the location of the Mozambique ocean is difficult to determine in the early Tonian due to 1664 overlapping latitudes between Azania and India (they are separated longitudinally by ~120° in our model). 1665 In our model there is no spreading in this ocean basin at this time, because Congo-Azania was latitudinally 1666 stable while India-South China moves towards the North Pole on a different longitude, accounted for by 1667 spreading in the Mirovoi Ocean. The spreading ridge in the Mawson Sea at ca. 900 Ma extends sufficiently 1668 south so that the ocean lithosphere generated here is subducted during closure of the Mozambique Ocean. 1669 We model active spreading in a clearly defined Mozambique Ocean beginning at 850 Ma, by which time 1670 India-South China had a similar longitude to Azania-Congo, making it easier to delineate the geographical 1671 extent of the ocean distinct from the Mawson Sea. Between 820 and 750 Ma the ocean basin closes rapidly, 1672 in order to fit palaeomagnetic constraints at 750 Ma that show India at ~60°N (Gregory et al., 2009; Torsvik 1673 et al., 2001) and Congo at 15°S (Meert et al., 1995) (placing Azania at the equator). 1674 1675 At 800 Ma, we model the birth of the Pacific/Panthalassic Ocean, here defined as the ocean basin separating 1676 Laurentia from North China, Australia and Antarctica, which most likely opened as Rodinia broke-up 1677 (although terrane migration across this ocean basin cannot be ruled out, e.g. Mulder et al. (2020). We note 1678 here that the pre-Mesozoic Pacific Ocean is already universally referred to as the Panthalassic Ocean. 1679 Consequently, we refer to the ocean separating North China-Australia-Antarctica and Laurentia between 1680 Rodinia and Gondwana times (ca. 800–520 Ma) as the proto-Pacific Ocean, the ocean surrounding 1681 Gondwana as the Panthalassic Ocean (‘Panthalassa’, ca. 520–200 Ma) and the ocean that has existed from 1682 200 Ma to present-day as the Pacific Ocean, noting that all three of these oceans essentially refer to the 1683 same ocean basin (in a geographical sense) that existed between North China-Australia-Antarctica and 1684 Laurentia. From 800 to 720 Ma we have a single ridge system separating Australia-Antarctica and 1685 Laurentia. This ridge produces a highly angular divergence, with spreading rates faster towards northern 1686 Australia and Canada than in Antarctica, resulting in a wider ocean basin in the north and narrower in the 1687 south. These variable spreading rates are required fit the ca. 750 Ma palaeomagnetic data, which require 1688 Australia to be ‘upright’ (same orientation as present-day) and perpendicular to Laurentia (Wingate and 1689 Giddings, 2000). 1690 1691 6.1.2 late Tonian–Cambrian (Rodinia breakup–Gondwana Assembly, 720–520 Ma) 1692 1693 By 720 Ma, continental motions around the Mirovoi Ocean are latitudinally stable with no polar excursions 1694 thus, the simplest way to account for the necessary subduction is with a stable triple junction. At this time, 1695 we show subduction along the western margin of the Mirovoi Ocean, outboard of Siberia in the Taimyr 1696 region, as well further south outboard of the WAC and Baltica where the Avalonian and Cadomian terranes 1697 were coalescing (Murphy et al., 1999; Murphy and Nance, 1989) and along the northern margin of the 1698 Sahara Metacraton in the east. This latter subduction zone is speculative since there are no known rocks of 1699 this age in the Sahara Metacraton (Blades et al., submitted). The Mirovoi ocean basin does not grow 1700 appreciably during the Ediacaran due to onset of subduction around its periphery, however, at 590 Ma 1701 Baltica begins moving north, resulting in the consumption of the Mirovoi Ocean basin and necessitating a 1702 change in mid-ocean ridge configuration. At this point we model a single ridge parallel to the Gondwanan 1703 margin, from Amazonia in the west towards India in the northeast. This ridge also accommodates the final 1704 motion of India as it collides with Congo. It is difficult to identify when the Mirovoi Ocean ceased to exist, 1705 but by ca. 490 Ma, expansion in the Proto-Tethyan ocean basin outboard of northwest Gondwana (e.g. Fig. 1706 12), as well as the easterly drift of Siberia amalgamates the Siberian subduction zones outboard of North 1707 China and Chu-Yili-Tianshan, probably indicating that the majority of Mirovoi crust produced during the 1708 Cryogenian and Ediacaran has been consumed. 1709 1710 An obvious issue with this discussion on the size of the Mirovoi is the longitudinal uncertainty between the 1711 position of Congo and Laurentia during the Ediacaran. This uncertainty is because the distance between 1712 these two cratons dictates the size of the Adamastor Ocean (between Kalahari and Congo), which grew at 1713 expense of the Mirovoi Ocean. Earliest evidence of subduction in the Damara belt exists from 650 Ma, with 1714 final closure occurring at 550 Ma. A conservative convergence rate of 40–60 km/Ma (roughly equivalent 1715 to present-day Pacific convergence rates) would make the ocean basin at 4000–6000 km wide, roughly 1716 equivalent to the present-day (this width is similar in our model to that of the Adamastor 1717 Ocean). However, faster divergence during Rodinia breakup would increase the size of this ocean basin, in 1718 turn reducing the size of the Mirovoi Ocean during the Ediacaran, similar to how the size of the modern 1719 Pacific Ocean would become smaller or larger depending on the changing size of the Atlantic Ocean. 1720 1721 The Mawson Sea remains very small in size during the Cryogenian due to the close proximity between 1722 India and Australia-Antarctica. There is no active ridge, instead a transform fault separates the two 1723 continents. There is little evidence of subduction-related magmatism on either cratonic Australia or India 1724 or this time (e.g. Halpin et al., 2017), suggesting the intervening lithosphere that eventually closed with 1725 Gondwana amalgamation could have also involved a more complex scenario of terrane accretion. The 1726 veracity of that statement is strongly dependent on the configuration and relative positioning of other 1727 terranes that are typically reconstructed to the north-western margin of Australia in the late Palaeozoic, such 1728 as Sibumasu and Indochina (Metcalfe, 2011) or other terranes preserved in Antarctica that are speculated 1729 to have rifted off the Indo-Antarctica and accreted to the western margin of the Mawson Craton (Daczko et 1730 al., 2018; Mulder et al., 2019). Neither set of terranes are yet considered explicitly in our model. By 520 1731 Ma Australia-Antarctica is sutured with India, closing any remnants of the Mawson Sea. 1732 1733 Both the Mozambique and NeoMozambique oceans close orthogonally from 720 to 550 Ma due to the 1734 continual southward motion of India towards Congo-SM. We do not model an active spreading ridge during 1735 this time, as subduction is only preserved on the African side of the collision (Collins and Pisarevsky, 2005). 1736 The presence of an earlier ridge does, however, suggest that at least two ridges (in our model they are 1737 extinct) were subducted during the . 1738 1739 The Proto-Pacific Ocean grows predominantly longitudinally during the Cryogenian–Cambrian. In our 1740 model, we show separation between Australia-Antarctica-North China and Laurentia using a single ridge 1741 system that propagates southwards, around southern Laurentia to eventually separate the 1742 and DML at ca. 700 Ma. We maintain a single ridge system until the Ediacaran, although when Kalahari 1743 begins drifting from Laurentia, we follow Merdith et al. (2017b) in inferring a ridge jump to re-align 1744 spreading between Australia and Laurentia with the incipient ridge separating Kalahari and Laurentia. This 1745 ridge jump also assists with reconciling the necessary motion of Australia to Ediacaran palaeomagnetic 1746 data, which require a ~35–45° counter-clockwise rotation from its present-day orientation (Schmidt and 1747 Williams, 2010). At 580 Ma we model a triple junction in the Proto-Pacific Ocean that coincides with the 1748 equatorial excursion of Baltica and cessation of triple junction spreading in the Mirovoi Ocean. The arms 1749 of this triple junction intersect sub-perpendicular to subduction outboard of Laurentia, a transform fault 1750 outboard of North China and another transform boundary outboard of Baltica. 1751 1752 Between 500 and 410 Ma, we refer to the ocean as the Panthalassic Ocean and for this time interval we 1753 extended the triple junction of YOU19 backwards through DOM16 and DOM18. We found at 500 Ma 1754 when Cuyania rifts off the promontory of Laurentia (Domeier, 2016), that the position of the triple junction 1755 and orientation of the ridges extended backwards from YOU19 was parallel to the direction of spreading 1756 separating Cuyania from Laurentia. While coincidental, given the arbitrary nature of pre-Mesozoic ocean 1757 plates, we find it useful to use the configuration at this time as a transition from the Proto-Pacific Ocean to 1758 the Panthalassic Ocean. 1759 1760 7 Conclusions 1761 1762 We present here the first continuous full-plate model from 1 Ga to present-day. The model traces the 1763 kinematic evolution of all cratonic crust and links the Neoproterozoic to the Phanerozoic, encompassing an 1764 entire supercontinent cycle, and enabling quantitative analysis of tectonic features for deep time. We present 1765 the model in a palaeomagnetic reference frame, including a new APWP for Gondwana from 540 to 320 Ma 1766 and a GAPWaP from 320 to 0 Ma. For the Neoproterozoic, the model uses a hybrid hierarchy, where 1767 relative plate motions are tied to a key plate, forming distinct nodes. This cluster-approach allows for the 1768 model to be iterated, constructed and modified in the future more easily in light of sparse palaeomagnetic 1769 data, but abundant geological data. Our revised Neoproterozoic model incorporates a late amalgamation of 1770 Rodinia with a novel configuration, in particular through the removal of India, South China and Tarim from 1771 the supercontinent. We incorporate major plate re-organisation events at ca. 850 Ma and again at ca. 750– 1772 700 Ma, corresponding to the counter-clockwise rotation of Congo-Sao Francisco-Sahara Metacraton 1773 relative to Rodinia and the initial closing of the Mozambique Ocean and coeval opening of the Proto-Pacific 1774 Ocean, respectively. Our model also includes preliminary interpretations of the Neoproterozoic history of 1775 many regional areas, such as terrane amalgamation outboard of India-South China, Hoggar and Avalonia, 1776 that then link coherently with their more established Phanerozoic histories. We reiterate that this model is 1777 a non-unique solution of global palaeogeography and tectonics for the Neoproterozoic but we hope it can 1778 provide a framework on which future studies can build upon. To facilitate this, we include in our 1779 supplementary material (SM3) the tectonic parameters of seafloor production and consumption as extracted 1780 from the model. 1781 1782 Because our model has continuous plate boundaries, it enables a range of new scientific experiments such 1783 as those seeking to link plate boundary processes to other aspects of the Earth system. This includes 1784 experiments related to the biosphere, hydrosphere and atmosphere investigating events surrounding 1785 oxygenation of Earth’s atmosphere, and radiation (e.g. (Gernon et al., 2016; 1786 Goddéris et al., 2017; He et al., 2019; Hoffman et al., 1998; Hoffman and Schrag, 2002; Lenton et al., 2016; 1787 Mills et al., 2019) and those studying the deep Earth (e.g. Heron et al., 2020). There are a number of 1788 limitations of this study, in particular, we do not address TPW in this model. Most previous studies looked 1789 at TPW purely from a palaeomagnetic framework, however the incorporation of geological data in the form 1790 of plate boundaries in this model (and others like it) open up opportunities for to analyse whole-lithospheric 1791 motion from other directions (e.g. Tetley et al., 2019). 1792 1793 Acknowledgements 1794 1795 The authors thank Sergei Pisarevsky for advice on palaeomagnetic data. ASM is currently supported by the 1796 Deep Energy Community of the Deep Carbon Observatory and the Richard Lounsbery Foundation. ASM, 1797 SEW, RDM and SZ were supported by Australian Research Council grant IH130200012. ASC and MLB 1798 are supported by Australian Research Council grants FT120100340 and LP160101353. MGT was 1799 supported by European Research Council Grant Agreement 617588 and Agence Nationale de la Recherche 1800 project ANR-18-ERC1-0005. JAM is supported by ARC grant FL160100168. AY and SA are supported 1801 by an Australian Government Research Training Program Scholarship. SZ was supported by Australian 1802 Research Council grant IH130200012, a University of Sydney Robinson Fellowship. SZ and RDM were 1803 supported by Alfred P. Sloan grants G-2017-9997 and G-2018-11296 through the Deep Carbon 1804 Observatory. JC and RDM were also supported by the AuScope National Collaborative Research 1805 Infrastructure System (NCRIS) program. pyGPlates and GPlates development is funded by the AuScope 1806 National Collaborative Research Infrastructure System (NCRIS) program. The authors are grateful for 1807 thorough reviews by D. J. J. van Hinsbergen, C. Scotese and B. Murphy, all of which greatly improved the 1808 focus and clarity of the manuscript. 1809 1810 1811 Table 1 Model Time (Ma) Scope Reference Frame LLSVPs Hierarchy Structure Palaeomagnetic Mantle Details SET12 200–0 Global No Yes TPW corrected palaeomagnetic data (200–100 Ma). No assumption. Branching. Hotspot motion (100–0 Ma). DT14 410–250 Global Yes Yes TPW corrected mantle reference from 410–250 Ma. Assumes stable and long-lived. Flat. MUL16 230–200 Global No Yes TPW corrected palaeomagnetic data (200–100 Ma). No assumption. Branching. Hotspot motion (100–0 Ma). MAT16 410–0 Global Yes Yes TPW corrected mantle reference from 410–0 Ma. Assumes stable and long-lived. Branching. DOM16 500–410 Gondwana-Laurentia-Baltica Yes Yes TPW corrected mantle reference from 500–410 Ma. Assumes stable and long-lived. Flat. MER17 1000–520 Global Yes No No assumption. Nodal. DOM18 500–410 Gondwana-Siberia-China Yes Yes Assumes stable and long-lived. Flat. TPW corrected mantle reference from 500–410 Ma. YOU19 410–0 Global Yes Yes TPW corrected mantle reference frame from 250–0 Ma. No assumption. Branching. MUL19 250–0 Global No Yes Optimisation method after Tetley et al. (2019). No assumption. Branching. This model 1000–0 Global Yes No No assumption. Nodal from 1000–410 Ma. Branching from 410–0 Ma. 1812 1813 Table 2

Key Rockunit OldAge YoungAge Glat Glon Plat Plon A95 Reference

Laurentia

L1 Gunbarrel Intrusions 780 776 45 -110 14.6 127 3.2 Harlan et al. (2008). combined

L2 800 750 41 -110 0.8 161.3 4.7 Weil et al. (2006).

L3 Galeros - Carbon 764 750 35.15 -111.8 -0.5 166 9.7 Weil et al. (2004); Eyster et al. Canyon (2019).

L4 Kwagunt Formation 759 743 36.15 -112 14.2 163.8 3.5 Eyster et al. (2019).

L5 Kwagunt Formation 748 736 36.15 -112 18.2 166 7 Weil et al. (2004). 2

L6 Franklin Dykes 727 712 75 -82 8.4 163.8 2.8 Denyszyn et al. (2009).

L7 Long Range Dykes 617 613 53.5 -57.5 -19 175.3 17.4 Murthy et al. (1992); Hodych et al. (2004); Age: Kamo and Gower (1994).

L8 Skinner Cove 554 548 50 -60 15 157 9 McCausland and Hodych (1998). Formation

L9** Andres Red Beds 423 393 41 -74 13 105 9 Miller and Kent (1988).

L10** Wabash Reef 423 415 40.85 -85.7 17 125 5.3 McCabe et al. (1985); Torsvik et al. (1996).

L11** 433 427 39 -79 19.1 128.3 5.8 French and Van der Voo (1979).

L12** Ringgold Gap 456 433 34.51 -85.06 24 146.6 7.7 Morrison and Ellwood (1986). Sediments

L13** Tablehead Group 470 456 48.33 -58.43 13.4 149.3 3.9 Hodych (1989); Hall and Evans mean (1988); Deutsch and Prasad (1987); Torsvik et al. (1996); Torsvik et al. (2012).

L14** St George Group 485 456 48.3 -59 17.5 152.4 4.3 Deutsch and Prasad (1987). Limestone

L15** Oneota Dolomite 485 470 43.41 -91.23 10.3 166.5 11.9 Jackson and Van der Voo (1985) . L16** Moore Hollow 500 490 31 -99 -0.6 163 8.5 Far and Gose (1991). sediments

L17** Morgan Creek 497 470 30.25 -98.5 10.6 158 9.7 Loucks an Elmore (1986).

L18** Point Peak 497 485 30.5 -99 5.2 165.8 6 Van der Voo et al. (1976).

L19** Taum Sauk 497 485 37.55 -90.31 -3.4 175.1 7.1 Dunn and Elmore (1985). Limestone

L20** Roywe dolomite 497 485 34.25 -97.11 12.6 157.3 4.3 Nick and Elmore (1990).

L21** Florida Mountains 497 485 32.05 - -5.4 168.7 10 Geissman et al. (1991). 107.37

L22** Tapeats 520 497 36.11 - 2.3 162.6 3.3 Elston and Bressler (1977). 113.99

L23** Mount Rigaud and 534 530 45.28 24.2 -11.9 184.5 6.2 McClausland et al. (2007). Chatham-Grenville

Baltica

B1 Southern Sweden 946 935 59 16 -0.9 240.7 6.7 Elming et al. (2014); Pisarevsky Dykes and Bylund (1998).

B2 Branton-Algo 927 905 58.5 6.5 5 249 3.9 Stearn and Piper, (1984); Age: Anortthosite Scherstén et al. (2000).

B3 Rogaland Igneous 883 855 58.5 6 -46 238 18.1 Walderhaug et al. (2007). Complex

B4 Hunnedalen Dykes 875 821 59 6.75 -41 222 10 Walderhaug et al. (1999).

B5 Egersund Dykes 619 613 58.5 6 -31.4 224.1 15.6 Walderhaug et al. (2007).

B6 Kurgashlya 570 560 53.3 57.5 -51 135 4.9 Lubnina et al. (2014). Formation

B7 Bakeevo Formation 570 560 54.5 58.2 -42 119 5.3 Lubnina et al. (2014).

B8 Winter coast 558 552 65.5 40 -25.3 132.5 2.8 Popov et al. (2002); Age: Martin et sediments al. (2000).

B9 Zolotitsa sediments 560 550 65.5 40 -31.7 112.9 2.4 Popov et al. (2005). I, B10 Verkhotina 560 550 65.5 40 -32.2 107.1 2 Popov et al. (2005). sediments

B11 Zolotitsa sediments 560 550 65.5 40 -28.3 109.9 3.8 Iglesia Llanos et al. (2005). II

B12 Zigan Formation 552 544 53.7 56.7 -16 138 3.7 Levashova et al. (2013).

B13 Swedish Limestone 467 458 58.3 13.9 3 35 13.4 Torsvik & Trench (1991). (1N)

B14 Swedish Limestone 480 470 58 13 30 55 9 Torsvik & Trench (1991).

B15 Swedish Limestone 480 470 59 15 18 46 5.1 Torsvik & Trench (1991). (1R)

B16 Narva 485 470 59 31 18 55 4 Khramov and Iosifidi (2009).

B17 St Petersburg 480 470 58 30 33 58 3.6 Smethurst et al., (1998). Limestone

B18+ Gotland Follingbo 430 420 57.5 18.5 21 164 6 Claesson (1979). Limestone

B19+ Dniestr Silurian 428 416 48.6 27 14.3 169.3 7.4 Jelenska et al. (2005). Lmst.

B20+ Gotland Medby 427 417 57.5 18.5 23 171 8 Claesson (1979). Limest.

B21+ Ringerike Sandst. 426 410 60 10.2 19 164 6.7 Douglass. (1988). Norway

B22+ Tiverskaya Series 419 411 48.6 27 0 149 13.3 Jelenska et al. (2015).

B23+ Ivaniev and Dniestr 419 393 48.7 26 -1 175 9.6 Lubnina et al (2007). Sediments

B24+ Devonian Seds. 416 406 48.7 26 -3.7 145.5 6.7 Smethurst and Khramov (1992). Podolia

B25+ Dniestrovskaya 416 407 48.6 27 2.3 158.4 7.4 Jelenska et al. (2005). Series

B26+ Zilair Sediments, 411 375 54 59 -2 161 3.1 Danukalov et al. (1983). Russia

B27+ Eifelian sedimentary 398 392 50 5 19 173 2.9 Minibaev and Sulutdinov (1991). rocks, Russia B28+ Bashkirea 398 385 54 59 -7 162 4.8 Danukalov et al. (1983). Sediments, Russia

B29+ Kola Dykes, Russia 390 370 68 33 11 147.6 11.1 Veselovsky and Arzamastsev (2011).

Siberia

S1 Ust-Kirba Formation 960 930 60 137.2 8.1 2.6 10.4 Popov et al. (2002).

S2 Kitoi Dykes 762 754 52 103 0.4 21.8 6.1 Pisarevsky et al. (2013).

S3 Kesyussa Formation 542 535 71 122.5 -37.6 165 5.2 Pisarevsky et al. (1997).

S4- Moyero River 459 439 68 104 -14 124 8 Gallet and pavlov (1996). sediments

S5- Angara River 460 450 58.5 99.8 -29.5 140.2 6.4 Pavlov et al. (2012). sediments*

S6- Kulumbe section 466 456 68 88.8 -24.1 152.4 3.3 Pavlov et al. (2008).

S7- Stolobovaya section 466 456 62.1 92.5 -22 158 4 Pavlov et al. (2008).

S8- Moyero River 468 458 68 104 -23 158 4 Gallet and Pavlov (1996). sediments

S9- Angara River 473 463 58.5 99.8 -35.2 153.2 3.6 Pavlov et al. (2012). sediments

S10- Moyero River 474 464 68 104 -30 157 4 Gallet and Pavlov (1996). sediments

S11- Guragir Formation 475 465 68 88.8 -30.9 152.7 3.2 Pavlov and Gallet(1998).

S12- Angara River 480 470 58.5 99.8 -36.4 158.2 6.5 Pavlov et al. (2012). sediments

S13- Moyero River 483 473 67.5 104 -33.9 151.7 2.2 Gallet and Pavlov (1996). sediments S14- Uigur and 488 478 68 88.8 -35.2 127.2 4.9 Pavlov and Gallet (1998). Nizhneiltyk Formations

S15- Moyero River 488 478 68 104 -40 138 9 Gallet and Pavlov (1996). sediments

S16- Kulumbinskaya 505 495 68 88.8 -36.1 130.7 6 Pavlov and Gallet (1998). Formation

S17- Moyero River 505 495 68 104 -37 138 9 Gallet and Pavlov (1996). sediments

S18 Yunkyulyabit- 512 502 70.9 122.6 -36.4 139.6 4.6 Pisarevsky et al. (1997). Yuryakh Formation

S19- Kulumbe River 512 502 68 88.4 -41.9 135.8 2.3 Pavlov and Gallet (1998). section

S20- Khorbusuonka 512 502 71.5 124 -43.7 140.5 2.6 Gallet et al. (2003). Amgan and Mayan seds.

S21- Nyuya and 437 427 60.6 116.3 -17.6 102 3.2 Powerman et al. (2013). River sediments

S22- Nyuya and Lena 438 428 60.7 116.3 -18.6 101.9 4.6 Shatsillo et al. (2007). River sediments

S23- Lena River 454 444 60.5 116.4 -21 109 17.3 Torsvik et al. (1995). sediments

S24- Nyuya River 456 446 60.6 116.3 -31.3 129.5 3.6 Powerman et al. (2013). sediments

S25- Kudrino section 466 456 57.7 107.99 -21.1 143.4 5 Pavlov et al. (2008).

S26- Krivaya Luka 469 459 59.7 118.1 -25.6 117.9 5.1 Iosifide et al. (1999). Formation

S27- Krivaya Luka 469 459 59.7 118.1 -28.2 127.1 2.5 Iosifide et al. (1999). Formation

S28- Krivolutsky Suite 470 460 57.6 107.8 -32.6 137 8.3 Rodionov et al. (2003).

S29- Lena River 473 463 59.8 118.1 -32 139 3.1 Torsvik et al. (1995). sediments S30- Lena River redbeds 475 465 60 114 -25 137 9 Rodionev et al. (1966).

S31- Surinsk Formation 485 475 58.3 109.61 -42.2 128.1 5.8 Surkis et al. (1999).

S32- Verkholensk 506 496 58.5 109.8 - 124 4.5 Rodionev et al. (1998). Formation 37.69

S33- Maya River 515 505 60 132 -45.8 115 5 Pavlovl et al. (2008). sediments

West Australia

WA1 Browne Formation 830 730 -26 126 44.5 141.7 6.8 Pisarevsky et al. (2007).

WA2 Hussar Formation 800 730 -26 126 62.2 85.8 10.3 Pisarevsky et al. (2007).

WA3 Mundine Dykes 758 752 -23 115.8 45.3 135.4 4.1 Wingate and Giddings (2000).

North Australia

NA1 Johnny's Creek 780 660 -24 133 15.8 83 13.5 Swanson-Hysell et al. (2012). Member

South Australia

SA1 Angepena Formation 660 640 -32 138 47.1 176.6 5.3 Williams and Schmidt (2015).

SA2 Yaltipena Formation 650 635 -31.5 139 44.2 172.7 8.2 Sohl et al. (1999).

SA3 Elatina Formation, 645 635 -32 137.5 49.9 164.4 13.5 Embleton and Williams (1986); MEAN Schmidt et al. (1991); Schmidt and Williams (1995); Sohl et al. (1999).

SA4 Nuccaleena 635 610 -31 139 32.3 170.8 2.9 Schmidt et al. (2009). Formation

SA5 Brachina Formation 620 590 -32.2 138 46 135.4 3.3 Schmidt and Williams (2010).

SA6 Bunyeroo Formation 590 570 -31.6 138.4 18.1 196.3 8.8 Schmidt and Williams (1996). SA7 Wonoka Formation 575 555 -31.3 138.6 5.2 210.5 4.9 Schmidt and Williams (2010).

North China

NC1 Huaibei Sills 890 913 876 34 117 -52.3 149.3 3.5 Fu et al. (2015). Ma

NC2 Wudaotang and 541 521 35.6 110.5 18.5 341.9 6.5 Huang et al. (1999). Xinji Fm

NC3 and 541 501 36 118 21.2 335.2 12.4 Zhao et al. (1992). Sediments

NC4 NE Sino-Korean 541 485 35.6 110.5 26.8 334.5 8.9 Gao et al. (1983) (recalculated by Massif Zhao et al. (1992)).

NC5- Zhangxia and 510 496 35.6 110.5 37 326.7 5.5 Huang et al. (1999). Xuzhuang Fms

NC6- Zhaogezhuang area 490 467 39.7 118.5 32.9 294.6 4.7 Yang et al. (2002). carbonates

NC7- Changshan and 496 485 35.6 110.5 31.7 329.6 5.4 Huang et al. (1999). Gushan Fms

NC8- Hebei and Shandong 499 461 36 118 28.8 310.9 12.3 Zhao et al. (1992). Sediments

NC9- Liangjiashan and 485 470 35.6 110.5 37.4 324.3 8.5 Huang et al. (1999). Lower Majiagou Fm

NC10- Jinghe 470 456 35.6 110.5 31.5 327.7 7 Huang et al. (1999).

NC11 Upper Majiagou 470 456 35.6 110.5 27.9 310.4 9.2 Yang et al. (1996). Formation

NC12- Tianjinshan and 480 464 37.2 105.5 31.8 326.5 9.5 Huang et al. (1999). Miboshan Formations

South China

SC1 Yanbian Dykes A 830 818 26.5 101.5 45.1 130.4 19 Niu et al. (2016). SC2 Xiaofeng Dykes 812 792 31 111 13.5 91 10.9 Li et al. (2004).

SC3* Yanbian Dykes B 814 798 26.5 101.5 14.1 32.5 20.4 Niu et al. (2016).

SC4 Liantuo Formation 735 705 30.8 111 9.9 160.3 4.6 Jing et al. (2015) and Evans et al. (2000), combined.

SC5 Nantuo Formation 641 631 28.5 110 7.5 161.6 5.9 Zhang et al. (2013); Zhang and Piper (1997).

SC6 Doushantuo 614 590 30.8 111 25.9 185.5 6.7 Zhang et al. (2015). Formation

SC7- Douposi Formation 510 496 32.1 106.2 -39.5 185.1 8.3 Lixin et al. (1998). (Wangcang)

SC8- Douposi Formation 510 496 32.4 106.3 -51.3 166 8.3 Yang et al. (2004). (Guangyuan)

SC9- Shiqian Redbeds 440 425 27.5 108 4.9 194.7 5.6 Opdyke et al. (1987).

SC10- Shiqian Redbeds 440 425 27.5 108 14.9 196.1 5.1 Huang et al. (2000).

SC11- Pagoda Formation 458 445 32.4 106.3 -45.8 191.3 3.6 Han et al. (2015).

Congo

C1 Luakela Volcanics A 770 757 -11.5 24.25 -40.2 122 14.1 Wingate et al. (2010).

C2 Mbozi Complex 773 713 -9.2 32.8 -46 145 6.7 Meert et al. (1995).

Sao Francisco

SF1 Bahia Dykes (N+R) 928 912 -14 -39 7.3 106.4 6.2 Evans et al. (2016).

Tarim

T1* Sugetbrak Formation 635 550 40.9 79.4 19.1 149.7 9.3 Zhan et al. (2007). T2* Zhamoketi Andesite 621 609 41.5 87.8 -4.9 146.7 3.9 Zhao et al. (2014).

T3 Lower Sugetbrak 640 615 41 79.5 -21.1 87.4 7 Wen et al. (2017). Formation

T4* Tereeken Cap 640 630 41.5 87.8 27.6 140.4 9.9 Zhao et al. (2014). Carbonate

T5 Qiaoenbrak 760 720 40.8 79.5 -6.3 17.5 9.1 Wen et al. (2013). Formation, Aksu

T6* Aksu Dykes 819 795 41.15 80.1 19 128 4.5 Chen et al. (2004).

T7 Baiyisi Volcanics 770 717 41.6 86.54 -17.7 14.2 6 Huang et al. (2005).

T8- Ordovician 485 470 41.3 83.4 -20.4 180.6 8.5 Fang et al. (1996). Limestones

T9- Aksu-Kalpin-Bachu 455 445 40.6 78.9 -40.7 183.3 4.8 Sun and Huang (2009). area sediments

T10 Red Sandstone 433 427 40.6 79.4 12.8 159.8 7.3 Zhao et al. (2014)(2014) average of three poles from Fang et al. (1996); Li et al. (1990); Fang et al. (1998).

India

I1 Malani Igneous 770 734 26 72 69.4 75.7 6.5 Meert et al. (2013). Suite Grand Mean

I2 Bhander and Rewa 650 530 26 78 47.3 212.7 5.8 McElhinny et al. (1978). formations

I3* Jodphur Group 570 520 27 73 1 164 6.7 Davis et al. (2014).

Seychelles

SE1 Mahe Dykes 753 747 -4.7 55.5 54.8 57.6 12.1 Torsvik et al. (2001).

Rio de la Plata

RP1 Sierra de las Animas 582 574 -34 -55.3 12.2 78.9 14.9 Rapalini et al. (2015). Complex

RP2 Sierra de los 600 500 -37.8 -59 15.1 72.6 12.4 Rapalini et al. (2006). Barrientos Redbeds

West African Craton

WAC1 Djebel Boho Volc. 547 526 30.4 -6.7 -27.3 207.1 14.9 Robert et al. (2017).

WAC2 Fajjoud and 572 551 30.2 -7.8 -21.9 211 15.6 Robert et al. (2017). Tadoughast Volc.

WAC3* Adrar-n-takoucht 577 564 30.4 -7.8 57.6 115.6 15.7 Robert et al. (2017). Volc

East Avalonia

EAV1#,1 Treffgarne volcanics 482 472 52 -5 -56 126 5.5 Trench et al. (1992).

EAV2# Stapeley volcanics 471 463 52.6 -3 -26.6 216.1 4.9 McCabe and Channell (1990).

EAV3# Builth igneous and 468 460 52.1 -3.3 -11 198 10 McCabe et al. (1992). sediments

EAV4# Tramore volcanics 461 449 52.1 -7.4 -11 198 8.5 Deutsch (1980); Trench and Torsvik (1991).

EAV5# Borrowdale 448 438 54.4 -3.2 -8.1 186.2 6.9 Millward and Evans (2003); volcanics Channell and McCabe (1992).

EAV6#,2 Midlands Minor 442 432 52.5 -1.5 -52.5 181.5 10.4 Vizan et al. (2003). Intrusives

EAV7#,2 Browgill redbeds 439 433 54.3 -2.5 -13.6 133.9 12.4 Channell et al. (1993).

EAV8#,2 Tortworth volcanics 437 431 51.7 -2.5 7 124 4.7 Torsvik et al. (1994).

EAV9#,2 Mendips volcanics 435 429 51.2 -2.5 -13 91 8.8 Torsvik et al. (1993).

Ganderia

GAN1# Bourinot Group 510 496 46.1 -60.4 -21 160 8.1 Johnson and Van der Vood (1985).

West Avalonia

WAV1# Nahant intrusives 490 488 42.4 -70.9 -34 140 3.9 Thompson et al. (2010).

WAV2# Dunn Point 463 457 45.8 -62.1 2 130 4.1 Johnson and Van Der Voo (1990); volcanics Hamilton and Murphy (2004).

WAV3# Cape St. Mary sills 443 439 46.8 -54 10 140 9 Hodych and Buchan (1998).

Carolinia

CAR1#,2 Cid Formation 450 445 35 -80.2 29.6 122.1 5.1 Vick et al. (1987). metasediments

CAR2#,2 Uwharrie and Cid 450 445 35.5 -80 20 80 14.2 Noel et al. (1988). Formation metaseds

Cuyania

CUY1# Pavon Formation 458 452 -34.6 -68.6 -3.6 166.4 6.6 Rapalini and Cingolani (2004). sediments

* pole not fit by model # from compilation of Domeier (2016) + from compilation of S. Pisarevsky (Pers. Comm.) - from compilation of Domeier (2018) ** from compilation of Torsvik et al. (2012) 1 inclination only used 2 no upper age constraint (fits inclination data) Glat, sample site latitude; Glon, sample site longitude; Plat, pole latitude; Plon, pole longitude; A95, 95% confidence ellipse.

1814 1815 1816 References 1817 1818 Ahmad, M., Munson, T.J., 2013. Geology and mineral resources of the Northern Territory. Northern 1819 Territory Geological Survey Special Publication 5. 1820 Alessio, B.L., Blades, M.L., Murray, G., Thorpe, B., Collins, A.S., Kelsey, D.E., Foden, J., Payne, J., Al- 1821 Khirbash, S., Jourdan, F., 2018. Origin and tectonic evolution of the NE basement of Oman: a 1822 window into the Neoproterozoic accretionary growth of India? Geol. Mag. 155, 1150–1174. 1823 https://doi.org/10.1017/S0016756817000061 1824 Alexeiev, D.V., Biske, G.S., Kröner, A., Tretyakov, A.A., Kovach, V.P., Rojas-Agramonte, Y., 2020. 1825 Ediacaran, Early Ordovician and early Silurian arcs in the South Tianshan orogen of Kyrgyzstan. 1826 J. Asian Earth Sci. 190, 104194. https://doi.org/10.1016/j.jseaes.2019.104194 1827 Alexeiev, D.V., Kröner, A., Kovach, V.P., Tretyakov, A.A., Rojas-Agramonte, Y., Degtyarev, K.E., 1828 Mikolaichuk, A.V., Wong, J., Kiselev, V.V., 2019. Evolution of Cambrian and Early Ordovician 1829 arcs in the Kyrgyz North Tianshan: Insights from U-Pb zircon ages and geochemical data. 1830 Gondwana Res. 66, 93–115. https://doi.org/10.1016/j.gr.2018.09.005 1831 Alexeiev, D.V., Ryazantsev, A.V., Kröner, A., Tretyakov, A.A., Xia, X., Liu, D.Y., 2011. Geochemical 1832 data and zircon ages for rocks in a high-pressure belt of Chu-Yili Mountains, southern 1833 Kazakhstan: Implications for the earliest stages of accretion in Kazakhstan and the Tianshan. J. 1834 Asian Earth Sci. 42, 805–820. https://doi.org/10.1016/j.jseaes.2010.09.004 1835 Archibald, D.B., Collins, A.S., Foden, J.D., 2017. Tonian arc magmatism in central Madagascar: the 1836 petrogenesis of the Imorona-Itsindro Suite. The Journal of. 1837 Archibald, D.B., Collins, A.S., Foden, J.D., Payne, J.L., Macey, P.H., Holden, P., Razakamanana, T., 1838 2018. Stenian–Tonian arc magmatism in west–central Madagascar: the genesis of the Dabolava 1839 Suite. J. Geol. Soc. London 175, 111–129. https://doi.org/10.1144/jgs2017-028 1840 Armistead, S.E., Collins, A.S., Merdith, A.S., Payne, J.L., Cox, G.M., Foden, J.D., Razakamanana, T., De 1841 Waele, B., 2019. Evolving marginal terranes during Neoproterozoic supercontinent 1842 reorganisation: constraints from the Bemarivo Domain in northern Madagascar. Tectonics. 1843 https://doi.org/10.1029/2018TC005384 1844 Armstrong, R.A., Master, S., Robb, L.J., 2005. Geochronology of the Nchanga , and constraints 1845 on the maximum age of the Katanga Supergroup, Zambian Copperbelt. J. Afr. Earth. Sci. 42, 32– 1846 40. https://doi.org/10.1016/j.jafrearsci.2005.08.012 1847 Arthaud, F., Matte, P., 1977. Late strike-slip faulting in and northern Africa: 1848 Result of a right-lateral shear zone between the Appalachians and the Urals. GSA Bulletin 88, 1849 1305–1320. 2.0.CO;2">https://doi.org/10.1130/0016-7606(1977)88<1305:LPSFIS>2.0.CO;2 1850 Ashwal, L.D., Solanki, A.M., Pandit, M.K., Corfu, F., Hendriks, B.W.H., Burke, K., Torsvik, T.H., 2013. 1851 Geochronology and geochemistry of Neoproterozoic Mt. Abu granitoids, NW India: Regional 1852 correlation and implications for Rodinia paleogeography. Precambrian Res. 236, 265–281. 1853 https://doi.org/10.1016/j.precamres.2013.07.018 1854 Austermann, J., Ben-Avraham, Z., , P., Heidbach, O., Schubert, G., Stock, J.M., 2011. Quantifying 1855 the forces needed for the rapid change of Pacific plate motion at 6Ma. Earth Planet. Sci. Lett. 307, 1856 289–297. https://doi.org/10.1016/j.epsl.2011.04.043 1857 Austermann, J., Kaye, B.T., Mitrovica, J.X., Huybers, P., 2014. A statistical analysis of the correlation 1858 between large igneous provinces and lower mantle seismic structure. Geophys. J. Int. 197, 1–9. 1859 https://doi.org/10.1093/gji/ggt500 1860 Barker, C.A., Secor, Jr., T., D., Pray, J.R., Wright, J.E., 1998. Age and Deformation of the Longtown 1861 Metagranite, South Carolina Piedmont: A Possible Constraint on the Origin of the Carolina 1862 Terrane. J. Geol. 106, 713–726. https://doi.org/10.1086/516055 1863 Bebout, G.E., 1995. The impact of subduction-zone metamorphism on mantle-ocean chemical cycling. 1864 Chem. Geol. 126, 191–218. https://doi.org/10.1016/0009-2541(95)00118-5 1865 Berry, R.F., Steele, D.A., Meffre, S., 2008. Proterozoic metamorphism in Tasmania: Implications for 1866 tectonic reconstructions. Precambrian Res. 166, 387–396. 1867 https://doi.org/10.1016/j.precamres.2007.05.004 1868 Bevier, M.L., Barr, S.M., White, C.E., Macdonald, A.S., 1993. U–Pb geochronologic constraints on the 1869 volcanic evolution of the Mira (Avalon) terrane, southeastern Cape Breton Island, Nova Scotia. 1870 Can. J. Earth Sci. 30, 1–10. https://doi.org/10.1139/e93-001 1871 Black, L.P., 1997. Dating Tasmania’s oldest geological events. Australian Geological Survey 1872 Organisation. 1873 Black, R., Latouche, L., Liégeois, J.P., Caby, R., Bertrand, J.M., 1994. Pan-African displaced terranes in 1874 the (central Sahara). Geology 22, 641–644. 1875 2.3.CO;2">https://doi.org/10.1130/0091-7613(1994)022<0641:PADTIT>2.3.CO;2 1876 Blades, M.L., Alessio, B.L., Collins, A.S., Foden, J., Payne, J.L., Glorie, S., Holden, P., Thorpe, B., Al- 1877 Khirbash, S., 2019a. Unravelling the Neoproterozoic accretionary history of Oman, using an 1878 array of isotopic systems in zircon. J. Geol. Soc. London. https://doi.org/10.1144/jgs2018-125 1879 Blades, M.L., Collins, A.S., Foden, J., Payne, J.L., Stüwe, K., Hassan, M., Abu-Alam, T., submitted. Age 1880 and Hafnium Isotope Evolution of Sudanese Butana and Chad Illuminates the Stenian to 1881 Ediacaran Evolution of the Southeast Sahara Metacraton. Precambrian Res. 1882 Blades, M.L., Collins, A.S., Foden, J., Payne, J.L., Xu, X., Alemu, T., Woldetinsae, G., Clark, C., Taylor, 1883 R.J.M., 2015. Age and hafnium isotopic evolution of the Didesa and Kemashi Domains, western 1884 Ethiopia. Precambrian Res. 270, 267–284. https://doi.org/10.1016/j.precamres.2015.09.018 1885 Blades, M.L., Foden, J., Collins, A.S., Alemu, T., Woldetinsae, G., 2019b. The origin of the ultramafic 1886 rocks of the Tulu Dimtu Belt, western Ethiopia--do they represent remnants of the Mozambique 1887 Ocean? Geol. Mag. 156, 62–82. 1888 Bleeker, W., 2003. The late record: a puzzle in ca. 35 pieces. Lithos 71, 99–134. 1889 https://doi.org/10.1016/j.lithos.2003.07.003 1890 Bodorkos, S., Crowley, J.L., Claoué-Long, J.C., Anderson, J.R., Magee, C.W., 2020. Precise U–Pb 1891 baddeleyite dating of the Derim Derim Dolerite, McArthur Basin, Northern Territory: old and 1892 new SHRIMP and ID-TIMS constraints. Aust. J. Earth Sci. 1–15. 1893 https://doi.org/10.1080/08120099.2020.1749929 1894 Bower, D.J., Gurnis, M., Seton, M., 2013. Lower mantle structure from paleogeographically constrained 1895 dynamic Earth models. Geochem. Geophys. Geosyst. 14, 44–63. 1896 https://doi.org/10.1029/2012GC004267 1897 Boyden, J.A., Müller, R.D., Gurnis, M., Torsvik, T.H., Clark, J.A., Turner, M., Ivey-Law, H., Watson, 1898 R.J., Cannon, J.S., 2011. Next-generation plate tectonic reconstructions, in: Keller, G.R., Baru, C. 1899 (Eds.), Geoinformatics: Cyberinfrastructure for the Solid Earth Sciences. Cambridge University 1900 Press, Cambridge, pp. 95–113. 1901 Brenner, A.R., Fu, R.R., Evans, D.A.D., Smirnov, A.V., Trubko, R., Rose, I.R., 2020. Paleomagnetic 1902 evidence for modern-like plate motion velocities at 3.2 Ga. Science Advances 6, eaaz8670. 1903 https://doi.org/10.1126/sciadv.aaz8670 1904 Brown, M., Johnson, T.E., Gardiner, N.J., 2020a. Plate Tectonics and the Archean Earth. Annu. Rev. 1905 Earth Planet. Sci. 48, 291–320. https://doi.org/10.1146/annurev-earth-081619-052705 1906 Brown, M., Kirkland, C.L., Johnson, T.E., 2020b. Evolution of geodynamics since the Archean: 1907 Significant change at the dawn of the Phanerozoic. Geology 48, 488–492. 1908 https://doi.org/10.1130/G47417.1 1909 Brune, S., Williams, S.E., Müller, R.D., 2017. Potential links between continental rifting, CO2 degassing 1910 and climate change through time. Nat. Geosci. 10, 941–946. https://doi.org/10.1038/s41561-017- 1911 0003-6 1912 Burke, K., Torsvik, T.H., 2004. Derivation of Large Igneous Provinces of the past 200 million years from 1913 long-term heterogeneities in the deep mantle. Earth Planet. Sci. Lett. 227, 531–538. 1914 https://doi.org/10.1016/j.epsl.2004.09.015 1915 Burrett, C., Khin Zaw, Meffre, S., Lai, C.K., Khositanont, S., Chaodumrong, P., Udchachon, M., Ekins, 1916 S., Halpin, J., 2014. The configuration of Greater Gondwana—Evidence from LA ICPMS, U–Pb 1917 geochronology of detrital zircons from the Palaeozoic and Mesozoic of Southeast Asia and China. 1918 Gondwana Res. 26, 31–51. https://doi.org/10.1016/j.gr.2013.05.020 1919 Burrett, C., Long, J., Stait, B., 1990. Early-Middle Palaeozoic biogeography of Asian terranes derived 1920 from Gondwana. Geological Society, London, Memoirs 12, 163–174. 1921 https://doi.org/10.1144/GSL.MEM.1990.012.01.14 1922 Caby, R., 2003. Terrane assembly and geodynamic evolution of central–western Hoggar: a synthesis. J. 1923 Afr. Earth. Sci. 37, 133–159. https://doi.org/10.1016/j.jafrearsci.2003.05.003 1924 Caby, R., Andreopoulos-Renaud, U., Gravelle, M., 1982. Cadre geologique et geochronologique U/Pb sur 1925 zircon des batholites precoces dans le segment pan-africain du Hoggar central (Algerie). Bulletin 1926 de la Société Géologique de France S7-XXIV, 677–684. https://doi.org/10.2113/gssgfbull.S7- 1927 XXIV.4.677 1928 Caby, R., Andreopoulos-Renaud, U., Pin, C., 1989. Late Proterozoic arc–continent and continent– 1929 continent collision in the pan-African trans-Saharan belt of Mali. Can. J. Earth Sci. 26, 1136– 1930 1146. https://doi.org/10.1139/e89-097 1931 Calver, C.R., 2011. Neoproterozoic glacial deposits of Tasmania, in: Arnaud, E., Halverson, G.P., 1932 Shields-Zhou, G. (Ed.), The Geological Record of Neoproterozoic Glaciations. Geological 1933 Society of London, pp. 649–658. https://doi.org/10.1144/M36.64 1934 Calver, C.R., Everard, J.L., Berry, R.F., Bottrill, R.S., Seymour, D.B., 2014. Proterozoic Tasmania. 1935 Geological Evolution of Tasmania: Geological Society of Australia Special Publication 24, 33– 1936 94. 1937 Calver, C.R., Meffre, S., Everard, J.L., 2013. Felsic Porphyry Sills in Surprise Bay Formation Near 1938 Currie, King Island, Dated at 775 Ma (LA-ICPMS, U-Pb on Zircon). Tasmanian Geological 1939 Survey. 1940 Cao, W., Williams, S., Flament, N., Zahirovic, S., Scotese, C., Dietmar Müller, R., 2019. 1941 Palaeolatitudinal distribution of lithologic indicators of climate in a palaeogeographic framework. 1942 Geol. Mag. 156, 331–354. https://doi.org/10.1017/S0016756818000110 1943 Cawood, P.A., Buchan, C., 2007. Linking accretionary orogenesis with supercontinent assembly. Earth- 1944 Sci. Rev. 82, 217–256. https://doi.org/10.1016/j.earscirev.2007.03.003 1945 Cawood, P.A., Hawkesworth, C.J., Dhuime, B., 2013a. The continental record and the generation of 1946 continental crust. GSA Bulletin 125, 14–32. https://doi.org/10.1130/B30722.1 1947 Cawood, P.A., Hawkesworth, C.J., Pisarevsky, S.A., Dhuime, B., Capitanio, F.A., Nebel, O., 2018a. 1948 Geological archive of the onset of plate tectonics. Philos. Trans. A Math. Phys. Eng. Sci. 376. 1949 https://doi.org/10.1098/rsta.2017.0405 1950 Cawood, P.A., Nemchin, A.A., Leverenz, A., Saeed, A., Balance, P.F., 1999. U/Pb dating of detrital 1951 zircons: Implications for the provenance record of Gondwana margin terranes. GSA Bulletin 111, 1952 1107–1119. 2.3.CO;2">https://doi.org/10.1130/0016-7606(1999)111<1107:UPDODZ>2.3.CO;2 1953 Cawood, P.A., Nemchin, A.A., Smith, M., Loewy, S., 2003. Source of the Dalradian Supergroup 1954 constrained by U–Pb dating of detrital zircon and implications for the East Laurentian margin. J. 1955 Geol. Soc. London 160, 231–246. https://doi.org/10.1144/0016-764902-039 1956 Cawood, P.A., Pisarevsky, S.A., 2006. Was Baltica right-way-up or upside-down in the Neoproterozoic? 1957 J. Geol. Soc. London 163, 753–759. https://doi.org/10.1144/0016-76492005-126 1958 Cawood, P.A., Strachan, R., Cutts, K., Kinny, P.D., Hand, M., Pisarevsky, S., 2010. Neoproterozoic 1959 orogeny along the margin of Rodinia: Valhalla orogen, North Atlantic. Geology 38, 99–102. 1960 https://doi.org/10.1130/G30450.1 1961 Cawood, P.A., Strachan, R.A., Pisarevsky, S.A., Gladkochub, D.P., Murphy, J.B., 2016. Linking 1962 collisional and accretionary orogens during Rodinia assembly and breakup: Implications for 1963 models of supercontinent cycles. Earth Planet. Sci. Lett. 449, 118–126. 1964 https://doi.org/10.1016/j.epsl.2016.05.049 1965 Cawood, P.A., Wang, W., Zhao, T., Xu, Y., Mulder, J.A., Pisarevsky, S.A., Zhang, L., Gan, C., He, H., 1966 Liu, H., Qi, L., Wang, Y., Yao, J., Zhao, G., Zhou, M.-F., Zi, J.-W., 2020. Deconstructing South 1967 China and consequences for reconstructing Nuna and Rodinia. Earth-Sci. Rev. 204, 103169. 1968 https://doi.org/10.1016/j.earscirev.2020.103169 1969 Cawood, P.A., Wang, Y., Xu, Y., Zhao, G., 2013b. Locating South China in Rodinia and Gondwana: A 1970 fragment of greater India lithosphere? Geology 41, 903–906. https://doi.org/10.1130/G34395.1 1971 Cawood, P.A., Zhao, G., Yao, J., Wang, W., Xu, Y., Wang, Y., 2018b. Reconstructing South China in 1972 Phanerozoic and Precambrian supercontinents. Earth-Sci. Rev. 186, 173–194. 1973 https://doi.org/10.1016/j.earscirev.2017.06.001 1974 Caxito, F. de A., Uhlein, A., Dantas, E.L., 2014a. The Afeição augen-gneiss Suite and the record of the 1975 Cariris Velhos Orogeny (1000–960 Ma) within the Riacho do Pontal fold belt, NE Brazil. J. 1976 South Amer. Earth Sci. 51, 12–27. https://doi.org/10.1016/j.jsames.2013.12.012 1977 Caxito, F. de A., Uhlein, A., Stevenson, R., Uhlein, G.J., 2014b. Neoproterozoic oceanic crust remnants 1978 in northeast Brazil. Geology 42, 387–390. https://doi.org/10.1130/G35479.1 1979 Caxito, F.A., Uhlein, A., Dantas, E.L., Stevenson, R., Salgado, S.S., Dussin, I.A., Sial, A. da N., 2016. A 1980 complete Wilson Cycle recorded within the Riacho do Pontal Orogen, NE Brazil: Implications for 1981 the Neoproterozoic evolution of the Borborema Province at the heart of West Gondwana. 1982 Precambrian Res. 282, 97–120. https://doi.org/10.1016/j.precamres.2016.07.001 1983 Cayley, R.A., 2011. Exotic crustal block accretion to the eastern Gondwanaland margin in the Late 1984 Cambrian--Tasmania, the Selwyn Block, and implications for the Cambrian--Silurian evolution of 1985 the Ross, Delamerian, and Lachlan orogens. Gondwana Res. 19, 628–649. 1986 Channell, J.E.T., McCabe, C., 1992. Palaeomagnetic data from the Borrowdale Volcanic Group: volcano- 1987 tectonics and Late Ordovician palaeolatitudes. J. Geol. Soc. London 149, 881–888. 1988 https://doi.org/10.1144/gsjgs.149.6.0881 1989 Channell, J.E.T., McCabe, C., Woodcock, N.H., 1993. Palaeomagnetic study of Llandovery (Lower 1990 Silurian) red beds in north-west England. Geophys. J. Int. 115, 1085–1094. 1991 Charvet, J., Shu, L., Laurent-Charvet, S., Wang, B., Faure, M., Cluzel, D., Chen, Y., De Jong, K., 2011. 1992 Palaeozoic tectonic evolution of the Tianshan belt, NW China. Sci. China Earth Sci. 54, 166–184. 1993 https://doi.org/10.1007/s11430-010-4138-1 1994 Chen, Y., Xu, B., Zhan, S., Li, Y., 2004. First mid-Neoproterozoic paleomagnetic results from the Tarim 1995 Basin (NW China) and their geodynamic implications. Precambrian Res. 133, 271–281. 1996 https://doi.org/10.1016/j.precamres.2004.05.002 1997 Claesson, C., 1979. Early Palaeozoic geomagnetism of Gotland. Geologiska Föreningen i Stockholm 1998 Förhandlingar 101, 149–155. https://doi.org/10.1080/11035897909452573 1999 Clennett, E.J., Sigloch, K., Mihalynuk, M.G., Seton, M., Henderson, M.A., Hosseini, K., 2000 Mohammadzaheri, A., Johnston, S.T., Müller, R.D., 2020. A Quantitative Tomotectonic Plate 2001 Reconstruction of Western North America and the Eastern Pacific Basin. Geochem. Geophys. 2002 Geosyst., Geological Association of Canada Special Paper 21, 2011‐11. 2003 https://doi.org/10.1029/2020GC009117 2004 Cocks, L.R.M., Fortey, R.A., 1997. A new Hirnantia faunafrom Thailand and the biogeography of the 2005 latest Ordovician of south-east Asia. Geobios Mem. Spec. 30, 117–126. 2006 https://doi.org/10.1016/S0016-6995(97)80017-4 2007 Cocks, L.R.M., Torsvik, T.H., 2007. Siberia, the wandering northern terrane, and its changing geography 2008 through the Palaeozoic. Earth-Sci. Rev. 82, 29–74. 2009 https://doi.org/10.1016/j.earscirev.2007.02.001 2010 Collins, A.S., 2006. Madagascar and the amalgamation of Central Gondwana. Gondwana Res. 9, 3–16. 2011 https://doi.org/10.1016/j.gr.2005.10.001 2012 Collins, A.S., 2003. Structure and age of the northern Leeuwin Complex, Western Australia: Constraints 2013 from field mapping and U–Pb isotopic analysis. Aust. J. Earth Sci. 50, 585–599. 2014 https://doi.org/10.1046/j.1440-0952.2003.01014.x 2015 Collins, A.S., Blades, M.L., Merdith, A.S., Foden, J.D., in revision. A late Tonian plate reorganization 2016 event revealed by a full-plate Proterozoic reconstruction. Communications Earth and 2017 Environment. 2018 Collins, A.S., Kröner, A., Fitzsimons, I.C.W., Razakamanana, T., 2003. Detrital footprint of the 2019 Mozambique ocean: U–Pb SHRIMP and Pb evaporation zircon geochronology of 2020 metasedimentary gneisses in eastern Madagascar. Tectonophysics 375, 77–99. 2021 https://doi.org/10.1016/S0040-1951(03)00334-2 2022 Collins, A.S., Pisarevsky, S.A., 2005. Amalgamating eastern Gondwana: The evolution of the Circum- 2023 Indian Orogens. Earth-Sci. Rev. 71, 229–270. https://doi.org/10.1016/j.earscirev.2005.02.004 2024 Collins, W.J., Belousova, E.A., Kemp, A.I.S., Murphy, J.B., 2011. Two contrasting Phanerozoic orogenic 2025 systems revealed by hafnium isotope data. Nat. Geosci. 4, 333–337. 2026 https://doi.org/10.1038/ngeo1127 2027 Colpron, M., Logan, J.M., Mortensen, J.K., 2002. U-Pb zircon age constraint for late Neoproterozoic 2028 rifting and initiation of the lower Paleozoic passive margin of western Laurentia. Can. J. Earth 2029 Sci. 39, 133–143. 2030 Compston, W., Wright, A.E., Toghill, P., 2002. Dating the Late Precambrian volcanicity of England and 2031 Wales. J. Geol. Soc. London 159, 323–339. https://doi.org/10.1144/0016-764901-010 2032 Corvino, A.F., Boger, S.D., Henjes-Kunst, F., Wilson, C.J.L., Fitzsimons, I.C.W., 2008. Superimposed 2033 tectonic events at 2450 Ma, 2100 Ma, 900 Ma and 500 Ma in the North Mawson Escarpment, 2034 Antarctic Prince Charles Mountains. Precambrian Res. 167, 281–302. 2035 Cox, G.M., Halverson, G.P., Denyszyn, S., Foden, J., Macdonald, F.A., 2018. Cryogenian magmatism 2036 along the north-western margin of Laurentia: Plume or rift? Precambrian Res. 319, 144–157. 2037 https://doi.org/10.1016/j.precamres.2017.09.025 2038 da Silva Filho, A.F., Guimarães, I.P., Van Schmus, W.R., 2002. Crustal Evolution of the Pernambuco- 2039 Alagoas Complex, Borborema Province, NE Brazil: Nd Isotopic Data from Neoproterozoic 2040 Granitoids. Gondwana Res. 5, 409–422. https://doi.org/10.1016/S1342-937X(05)70732-2 2041 Daczko, N.R., Halpin, J.A., Fitzsimons, I.C.W., Whittaker, J.M., 2018. A cryptic Gondwana-forming 2042 orogen located in Antarctica. Sci. Rep. 8, 8371. https://doi.org/10.1038/s41598-018-26530-1 2043 Dallmeyer, R.D., 1989. Contrasting accreted terranes in the southern Appalachian Orogen, basement 2044 beneath the Atlantic and Gulf Coastal Plains, and West African orogens. Precambrian Res. 42, 2045 387–409. https://doi.org/10.1016/0301-9268(89)90021-1 2046 Dalziel, I.W., 1992. On the organization of American plates in the Neoproterozoic and the breakout of 2047 Laurentia. GSA Today 2. 2048 Dalziel, I.W.D., 1997. Neoproterozoic-Paleozoic geography and tectonics: Review, hypothesis, 2049 environmental speculation. GSA Bulletin 109, 16–42. 2.3.CO;2">https://doi.org/10.1130/0016- 2050 7606(1997)109<0016:ONPGAT>2.3.CO;2 2051 Dalziel, I.W.D., 1991. Pacific margins of Laurentia and East Antarctica-Australia as a conjugate rift pair: 2052 Evidence and implications for an Eocambrian supercontinent. Geology 19, 598–601. 2053 Dalziel, I.W.D., Dewey, J.F., 2019. The classic Wilson cycle revisited. Geological Society, London, 2054 Special Publications 470, 19–38. https://doi.org/10.1144/SP470.1 2055 Danukalov, N.F., Kondruchina, L.S., Chernikov, A.P., 1983. Paleozoic of the South and 2056 Central Urals. BF AN SSSR, Ufa. 2057 Davies, D.R., Goes, S., Sambridge, M., 2015. On the relationship between volcanic hotspot locations, the 2058 reconstructed eruption sites of large igneous provinces and deep mantle seismic structure. Earth 2059 Planet. Sci. Lett. 411, 121–130. https://doi.org/10.1016/j.epsl.2014.11.052 2060 Davies, R.G., Crawford, A.R., 1971. Petrography and age of the rocks of Bulland Hill, Kirana Hills, 2061 Sarghoda District, West Pakistan. Geol. Mag. 108, 235–246. 2062 https://doi.org/10.1017/S001675680005158X 2063 Davis, J.K., Meert, J.G., Pandit, M.K., 2014. Paleomagnetic analysis of the Marwar Supergroup, 2064 Rajasthan, India and proposed interbasinal correlations. J. Asian Earth Sci. 91, 339–351. 2065 https://doi.org/10.1016/j.jseaes.2013.09.027 2066 Deb, M., Thorpe, R.I., Krstic, D., Corfu, F., Davis, D.W., 2001. Zircon U–Pb and galena Pb isotope 2067 evidence for an approximate 1.0 Ga terrane constituting the western margin of the Aravalli–Delhi 2068 orogenic belt, northwestern India. Precambrian Res. 108, 195–213. 2069 https://doi.org/10.1016/S0301-9268(01)00134-6 2070 Degtyarev, K., Yakubchuk, A., Tretyakov, A., Kotov, A., Kovach, V., 2017. Precambrian geology of the 2071 Kazakh Uplands and Tien Shan: An overview. Gondwana Res. 47, 44–75. 2072 https://doi.org/10.1016/j.gr.2016.12.014 2073 Degtyarev, K.E., 2011. Tectonic evolution of Early Paleozoic island-arc systems and continental crust 2074 formation in the Caledonides of Kazakhstan and the North Tien Shan. Geotectonics/Geotektonika 2075 45, 23–50. https://doi.org/10.1134/S0016852111010031 2076 Degtyarev, K.E., Shatagin, K.N., Kotov, A.B., Sal’Nikova, E.B., Luchitskaya, M.V., Tret’yakov, A.A., 2077 Yakovleva, S.Z., 2008. Late Precambrian volcanoplutonic association of the Aktau-Dzhungar 2078 massif, Central Kazakhstan: Structural position and age, in: Doklady Earth Sciences. Springer 2079 Nature BV, p. 879. 2080 Degtyarev, K.E., Tret’yakov, A.A., Ryazantsev, A.V., Kotov, A.B., Sal’nikova, E.B., Aleksandrov, P.A., 2081 Anisimova, I.V., 2011. Stenian granitoids of the west Kyrgyz Ridge (North Tien Shan): Position, 2082 structure, and age determination. Dokl. Earth Sci. 441, 1484–1488. 2083 https://doi.org/10.1134/S1028334X11110134 2084 Denyszyn, S.W., Halls, H.C., Davis, D.W., Evans, D.A.D., 2009. Paleomagnetism and U–Pb 2085 geochronology of Franklin dykes in High Arctic Canada and Greenland: a revised age and 2086 paleomagnetic pole constraining block rotations in the Nares Strait region. Can. J. Earth Sci. 46, 2087 689–705. https://doi.org/10.1139/E09-042 2088 Deutsch, E.R., 1980. Magnetism of the Mid-Ordovician Tramore Volcanics, SE Ireland, and the Question 2089 of a Wide Proto-Atlantic Ocean. J. Geomagn. Geoelectr. 32, SIII77–SIII98. 2090 https://doi.org/10.5636/jgg.32.Supplement3_SIII77 2091 Deutsch, E.R., Prasad, J.N., 1987. Ordovician paleomagnetic results from the St. George and Table Head 2092 carbonates of western . Can. J. Earth Sci. 24, 1785–1796. 2093 https://doi.org/10.1139/e87-170 2094 Direen, N.G., Crawford, A.J., 2003. Fossil seaward-dipping reflector sequences preserved in southeastern 2095 Australia: a 600 Ma volcanic passive margin in eastern Gondwanaland. J. Geol. Soc. London 160, 2096 985–990. https://doi.org/10.1144/0016-764903-010 2097 Doig, R., Murphy, J.B., Nance, R.D., 1993. Tectonic significance of the Late Proterozoic Economy River 2098 gneiss, Cobequid Highlands, Avalon Composite Terrane, Nova Scotia. Can. J. Earth Sci. 30, 474– 2099 479. https://doi.org/10.1139/e93-035 2100 Domeier, M., 2018. Early Paleozoic tectonics of Asia: Towards a full-plate model. Geoscience Frontiers 2101 9, 789–862. https://doi.org/10.1016/j.gsf.2017.11.012 2102 Domeier, M., 2016. A plate tectonic scenario for the Iapetus and Rheic oceans. Gondwana Res. 36, 275– 2103 295. https://doi.org/10.1016/j.gr.2015.08.003 2104 Domeier, M., Font, E., Youbi, N., Davies, J., Nemkin, S., Van der Voo, R., Perrot, M., Benabbou, M., 2105 Boumehdi, M.A., Torsvik, T.H., 2020. On the Early Permian shape of Pangea from 2106 paleomagnetism at its core. Gondwana Res. https://doi.org/10.1016/j.gr.2020.11.005 2107 Domeier, M., Shephard, G.E., Jakob, J., Gaina, C., Doubrovine, P.V., Torsvik, T.H., 2017. Intraoceanic 2108 subduction spanned the Pacific in the -. Sci Adv 3. 2109 https://doi.org/10.1126/sciadv.aao2303 2110 Domeier, M., Torsvik, T.H., 2017. Full-plate modelling in pre-Jurassic time. Geol. Mag. 1–20. 2111 https://doi.org/10.1017/S0016756817001005 2112 Domeier, M., Torsvik, T.H., 2014. Plate tectonics in the late Paleozoic. Geoscience Frontiers 5, 303–350. 2113 https://doi.org/10.1016/j.gsf.2014.01.002 2114 Domeier, M., Van der Voo, R., Torsvik, T.H., 2012. Paleomagnetism and Pangea: The road to 2115 reconciliation. Tectonophysics 514–517, 14–43. https://doi.org/10.1016/j.tecto.2011.10.021 2116 Dong, Y., Santosh, M., 2016. Tectonic architecture and multiple orogeny of the Qinling Orogenic Belt, 2117 Central China. Gondwana Res. 29, 1–40. https://doi.org/10.1016/j.gr.2015.06.009 2118 Dong, Y., Yang, Z., Liu, X., Zhang, X., He, D., Li, W., Zhang, F., Sun, S., Zhang, H., Zhang, G., 2014. 2119 Neoproterozoic amalgamation of the Northern Qinling terrain to the : 2120 Constraints from geochronology and geochemistry of the Kuanping ophiolite. Precambrian Res. 2121 255, 77–95. https://doi.org/10.1016/j.precamres.2014.09.008 2122 dos Santos, E.J., Van Schmus, W.R., Kozuch, M., Neves, B.B. de B., 2010. The Cariris Velhos tectonic 2123 event in Northeast Brazil. J. South Amer. Earth Sci. 29, 61–76. 2124 https://doi.org/10.1016/j.jsames.2009.07.003 2125 Doubrovine, P.V., Steinberger, B., Torsvik, T.H., 2016. A failure to reject: Testing the correlation 2126 between large igneous provinces and deep mantle structures with EDF statistics. Geochem. 2127 Geophys. Geosyst. 17, 1130–1163. 2128 Doucet, L.S., Li, Z.-X., Gamal El Dien, H., Pourteau, A., Murphy, J.B., Collins, W.J., Mattielli, N., 2129 Olierook, H.K.H., Spencer, C.J., Mitchell, R.N., 2020. Distinct formation history for deep-mantle 2130 domains reflected in geochemical differences. Nat. Geosci. 13, 511–515. 2131 https://doi.org/10.1038/s41561-020-0599-9 2132 Douglass, D.N., 1988. Paleomagnetics of Ringerike Old Red Sandstone and related rocks, southern 2133 Norway: implications for pre-Carboniferous separation of Baltica and British terranes. 2134 Tectonophysics 148, 11–27. https://doi.org/10.1016/0040-1951(88)90157-6 2135 Dunn, W.J., Elmore, R.D., 1985. Paleomagnetic and petrographic investigation of the Taum Sauk 2136 Limestone, southeast Missouri. J. Geophys. Res., Geodyn. Ser. 90, 11469. 2137 https://doi.org/10.1029/JB090iB13p11469 2138 Dutkiewicz, A., Dietmar Müller, R., Cannon, J., Vaughan, S., Zahirovic, S., 2019. Sequestration and 2139 subduction of deep-sea carbonate in the global ocean since the Early Cretaceous. Geology 47, 91– 2140 94. https://doi.org/10.1130/G45424.1 2141 Elburg, M.A., Andersen, T., Jacobs, J., Läufer, A., Ruppel, A., Krohne, N., Damaske, D., 2015. One 2142 Hundred Fifty Million Years of Intrusive Activity in the Sør Rondane Mountains (East 2143 Antarctica): Implications for Gondwana Assembly. J. Geol. https://doi.org/10.1086/684052 2144 Elming, S.-Å., Pisarevsky, S.A., Layer, P., Bylund, G., 2014. A palaeomagnetic and 40Ar/39Ar study of 2145 mafic dykes in southern Sweden: A new Early Neoproterozoic key-pole for the and 2146 implications for Sveconorwegian and Grenville loops. Precambrian Res. 244, 192–206. 2147 https://doi.org/10.1016/j.precamres.2013.12.007 2148 Elston, D.P., Bressler, S.L., 1977. Paleomagnetic poles and polarity zonation from Cambrian and 2149 Devonian strata of Arizona. Earth Planet. Sci. Lett. 36, 423–433. https://doi.org/10.1016/0012- 2150 821X(77)90067-X 2151 Embleton, B.J.J., Williams, G.E., 1986. Low palaeolatitude of deposition for late Precambrian periglacial 2152 varvites in South Australia: implications for palaeoclimatology. Earth Planet. Sci. Lett. 79, 419– 2153 430. https://doi.org/10.1016/0012-821X(86)90197-4 2154 England, P., Molnar, P., 1990. Surface uplift, uplift of rocks, and exhumation of rocks. Geology 18, 2155 1173–1177. 2.3.CO;2">https://doi.org/10.1130/0091-7613(1990)018<1173:SUUORA>2.3.CO;2 2156 Evans, D., 2003. True polar wander and supercontinents. Tectonophysics. https://doi.org/10.1016/S0040- 2157 1951(02)00642-X 2158 Evans, D.A.D., 2013. Reconstructing pre-Pangean supercontinents. Bull. Am. Assoc. Hist. Nurs. 2159 Evans, D.A.D., 2009. The palaeomagnetically viable, long-lived and all-inclusive Rodinia supercontinent 2160 reconstruction. Geological Society, London, Special. 2161 Evans, D.A.D., Li, Z.X., Kirschvink, J.L., Wingate, M.T.D., 2000. A high-quality mid-Neoproterozoic 2162 paleomagnetic pole from South China, with implications for ice ages and the breakup 2163 configuration of Rodinia. Precambrian Res. 100, 313–334. https://doi.org/10.1016/S0301- 2164 9268(99)00079-0 2165 Evans, D.A.D., Trindade, R.I.F., Catelani, E.L., D’Agrella-Filho, M.S., Heaman, L.M., Oliveira, E.P., 2166 Söderlund, U., Ernst, R.E., Smirnov, A.V., Salminen, J.M., 2016. Return to Rodinia? Moderate to 2167 high palaeolatitude of the São Francisco/Congo craton at 920 Ma. Geological Society, London, 2168 Special Publications 424, 167–190. https://doi.org/10.1144/SP424.1 2169 Eyster, A., Weiss, B.P., Karlstrom, K., Macdonald, F.A., 2019. Paleomagnetism of the Chuar Group and 2170 evaluation of the late Tonian Laurentian apparent polar wander path with implications for the 2171 makeup and breakup of Rodinia. GSA Bulletin. https://doi.org/10.1130/B32012.1 2172 Fang, D., Wang, P., Shen, Z., Tan, X., 1998. Cenozoic paleomagnetic results and phanerozoic apparent 2173 polar wandering path of Tarim Block. Sci. China Ser. D Earth Sci. 2174 Fang, D.-J., Jin, G.-H., Jiang, L.-P., Wang, P.-Y., Wang, Z.-L., 1996. Paleozoic paleomagnetic results and 2175 the tectonic significance of Tarim Plate. Acta Geophysica Sinica 39, 532–542. 2176 Farr, M.R., Gose, W.A., 1991. Paleomagnetism of the Cambrian Moore Hollow Group, Texas: Evidence 2177 for a primary magnetization carried by detrital magnetite. J. Geophys. Res. 96, 9895. 2178 https://doi.org/10.1029/91JB00337 2179 Fezaa, N., Liégeois, J.-P., Abdallah, N., Cherfouh, E.H., De Waele, B., Bruguier, O., Ouabadi, A., 2010. 2180 Late Ediacaran geological evolution (575–555 Ma) of the Djanet Terrane, Eastern Hoggar, 2181 , evidence for a Murzukian intracontinental episode. Precambrian Res. 180, 299–327. 2182 Fioretti, A.M., Black, L.P., Foden, J., Visonà, D., 2005. Grenville-age magmatism at the South Tasman 2183 Rise (Australia): A new piercing point for the reconstruction of Rodinia. Geology 33, 769–772. 2184 https://doi.org/10.1130/G21671.1 2185 Fitzsimons, I.C.W., 2003. Proterozoic basement provinces of southern and southwestern Australia, and 2186 their correlation with Antarctica. Geological Society, London, Special Publications 206, 93–130. 2187 https://doi.org/10.1144/GSL.SP.2003.206.01.07 2188 Flament, N., Williams, S., Müller, R.D., Gurnis, M., Bower, D.J., 2017. Origin and evolution of the deep 2189 thermochemical structure beneath Eurasia. Nat. Commun. 8, 14164. 2190 https://doi.org/10.1038/ncomms14164 2191 French, A.N., Van der Voo, R., 1979. The magnetization of the Rose Hill Formation at the classical site 2192 of Graham’s Fold Test. J. Geophys. Res. 84, 7688–7696. 2193 https://doi.org/10.1029/JB084iB13p07688 2194 Fu, X., Zhang, S., Li, Haiyan, Ding, J., Li, Huaikun, Yang, T., Wu, H., Yuan, H., Lv, J., 2015. New 2195 paleomagnetic results from the Huaibei Group and Neoproterozoic mafic sills in the North China 2196 Craton and their paleogeographic implications. Precambrian Res. 269, 90–106. 2197 https://doi.org/10.1016/j.precamres.2015.08.013 2198 Furnes, H., de Wit, M., Dilek, Y., 2014. Four billion years of ophiolites reveal secular trends in oceanic 2199 crust formation. Geoscience Frontiers 5, 571–603. https://doi.org/10.1016/j.gsf.2014.02.002 2200 Gallet, Y., Pavlov, V., 1996. of the Moyero River Section (North-Western Siberia): 2201 Constraints On Geomagnetic Reversal Frequency During the Early Palaeozoic. Geophys. J. Int. 2202 125, 95–105. https://doi.org/10.1111/j.1365-246X.1996.tb06536.x 2203 Gallet, Y., Pavlov, V., Courtillot, V., 2003. Magnetic reversal frequency and apparent polar wander of the 2204 Siberian platform in the earliest Palaeozoic, inferred from the Khorbusuonka river section 2205 (northeastern Siberia). Geophys. J. Int. 154, 829–840. 2206 Ganade de Araujo, C.E., Cordani, U.G., Weinberg, R.F., Basei, M.A.S., Armstrong, R., Sato, K., 2014. 2207 Tracing Neoproterozoic subduction in the Borborema Province (NE-Brazil): Clues from U-Pb 2208 geochronology and Sr-Nd-Hf-O isotopes on granitoids and migmatites. Lithos 202–203, 167– 2209 189. https://doi.org/10.1016/j.lithos.2014.05.015 2210 Gao, J., Wang, X.-S., Klemd, R., Jiang, T., Qian, Q., Mu, L.-X., Ma, Y.-Z., 2015. Record of assembly and 2211 breakup of Rodinia in the Southwestern Altaids: Evidence from Neoproterozoic magmatism in 2212 the Chinese Western Tianshan Orogen. J. Asian Earth Sci. 113, 173–193. 2213 https://doi.org/10.1016/j.jseaes.2015.02.002 2214 Gao, R.F., Huang, H.L., Zhu, Z.W., Liu, H.S., Fan, Y.Q., Qing, X.J., 1983. The study of paleomagnetism 2215 in northeastern Sino-Korean massif during Pre-late Paleozoic. Contr. Project of Plate Tectonics in 2216 North China 1, 265–274. 2217 Garnero, E.J., Lay, T., McNamara, A., 2007. Implications of lower-mantle structural heterogeneity for the 2218 existence and nature of whole-mantle plumes. Special Papers-Geological Society of America 430, 2219 79. 2220 Ge, R., Zhu, W., Wilde, S.A., He, J., Cui, X., Wang, X., Bihai, Z., 2014. Neoproterozoic to Paleozoic 2221 long-lived accretionary orogeny in the northern Tarim Craton: accretionary orogeny in northern 2222 Tarim. Tectonics 33, 302–329. https://doi.org/10.1002/2013TC003501 2223 Geissman, J.W., Jackson, M., Harlan, S.S., Van Der Voo, R., 1991. Paleomagnetism of Latest Cambrian- 2224 Early Ordovician and Latest Cretaceous-Early rocks of the Florida Mountains, southwest 2225 New Mexico. J. Geophys. Res. [Solid Earth] 96, 6053–6071. 2226 Gernon, T.M., Hincks, T.K., Tyrrell, T., Rohling, E.J., Palmer, M.R., 2016. Snowball Earth ocean 2227 chemistry driven by extensive ridge volcanism during Rodinia breakup. Nat. Geosci. 9, 242. 2228 https://doi.org/10.1038/ngeo2632 2229 Gerya, T., 2014. Precambrian geodynamics: Concepts and models. Gondwana Res. 25, 442–463. 2230 https://doi.org/10.1016/j.gr.2012.11.008 2231 Goddéris, Y., Le Hir, G., Macouin, M., Donnadieu, Y., Hubert-Théou, L., Dera, G., Aretz, M., Fluteau, 2232 F., Li, Z.X., Halverson, G.P., 2017. Paleogeographic forcing of the strontium isotopic cycle in the 2233 Neoproterozoic. Gondwana Res. 42, 151–162. https://doi.org/10.1016/j.gr.2016.09.013 2234 Goodge, J.W., Fanning, C.M., Fisher, C.M., Vervoort, J.D., 2017. Proterozoic crustal evolution of central 2235 East Antarctica: Age and isotopic evidence from glacial igneous clasts, and links with Australia 2236 and Laurentia. Precambrian Res. 299, 151–176. https://doi.org/10.1016/j.precamres.2017.07.026 2237 Gordon, R.G., Cox, A., O’Hare, S., 1984. Paleomagnetic Euler poles and the apparent polar wander and 2238 absolute motion of North America since the Carboniferous. Tectonics 3, 499–537. 2239 https://doi.org/10.1029/TC003i005p00499 2240 Granot, R., 2016. Palaeozoic oceanic crust preserved beneath the eastern Mediterranean. Nat. Geosci. 9, 2241 701–705. https://doi.org/10.1038/ngeo2784 2242 Gregory, L.C., Meert, J.G., Bingen, B., Pandit, M.K., Torsvik, T.H., 2009. Paleomagnetism and 2243 geochronology of the Malani Igneous Suite, Northwest India: Implications for the configuration 2244 of Rodinia and the assembly of Gondwana. Precambrian Res. 170, 13–26. 2245 https://doi.org/10.1016/j.precamres.2008.11.004 2246 Guimarães, I.P., Van Schmus, W.R., de Brito Neves, B.B., Bretas Bittar, S.M., Silva Filho, A.F., 2247 Armstrong, R., 2012. U–Pb zircon ages of orthogneisses and supracrustal rocks of the Cariris 2248 Velhos belt: Onset of Neoproterozoic rifting in the Borborema Province, NE Brazil. Precambrian 2249 Res. 192–195, 52–77. https://doi.org/10.1016/j.precamres.2011.10.008 2250 Gurnis, M., Turner, M., Zahirovic, S., DiCaprio, L., Spasojevic, S., Müller, R.D., Boyden, J., Seton, M., 2251 Manea, V.C., Bower, D.J., 2012. Plate tectonic reconstructions with continuously closing plates. 2252 Comput. Geosci. 38, 35–42. https://doi.org/10.1016/j.cageo.2011.04.014 2253 Gurnis, M., Yang, T., Cannon, J., Turner, M., Williams, S., Flament, N., Müller, R.D., 2018. Global 2254 tectonic reconstructions with continuously deforming and evolving rigid plates. Comput. Geosci. 2255 116, 32–41. https://doi.org/10.1016/j.cageo.2018.04.007 2256 Guynn, J., Kapp, P., Gehrels, G.E., Ding, L., 2012. U–Pb geochronology of basement rocks in central 2257 and paleogeographic implications. J. Asian Earth Sci. 43, 23–50. 2258 https://doi.org/10.1016/j.jseaes.2011.09.003 2259 Guynn, J.H., Kapp, P., Pullen, A., Heizler, M., Gehrels, G., Ding, L., 2006. Tibetan basement rocks near 2260 Amdo reveal “missing” Mesozoic tectonism along the , central Tibet. Geology 34, 2261 505–508. https://doi.org/10.1130/G22453.1 2262 Hadj-Kaddour, Z., Liégeois, J.-P., Demaiffe, D., Caby, R., 1998. The alkaline–peralkaline granitic post- 2263 collisional Tin Zebane dyke swarm (Pan-African Tuareg shield, Algeria): prevalent mantle 2264 signature and late agpaitic differentiation. Lithos 45, 223–243. https://doi.org/10.1016/S0024- 2265 4937(98)00033-4 2266 Haldar, S.K., Deb, M., 2001. Geology and mineralization of Rajpura-Dariba lead- belt, Rajasthan, in: 2267 Deb, M., Goodfellow, W.D. (Eds.), Sediment-Hosted Lead-Zink Deposit Modeling Program. 2268 Delhi-Udaipur, Elsevier, New Dehli, pp. 177–187. 2269 Hall, S.A., Evans, I., 1988. Paleomagnetic study of the Ordovician Table Head Group, Port au Port 2270 Peninsula, Newfoundland. Can. J. Earth Sci. 25, 1407–1419. https://doi.org/10.1139/e88-135 2271 Halpin, J.A., Daczko, N.R., Kobler, M.E., Whittaker, J.M., 2017. Strike-slip tectonics during the 2272 Neoproterozoic–Cambrian assembly of East Gondwana: Evidence from a newly discovered 2273 microcontinent in the …. Gondwana Res. 2274 Halpin, J.A., Jensen, T., McGoldrick, P., Meffre, S., Berry, R.F., Everard, J.L., Calver, C.R., Thompson, 2275 J., Goemann, K., Whittaker, J.M., 2014. Authigenic monazite and detrital zircon dating from the 2276 Proterozoic Rocky Cape Group, Tasmania: Links to the Belt-Purcell Supergroup, North America. 2277 Precambrian Res. 250, 50–67. https://doi.org/10.1016/j.precamres.2014.05.025 2278 Hamilton, M.A., Murphy, J.B., 2004. Tectonic significance of a Llanvirn age for the Dunn Point volcanic 2279 rocks, Avalon terrane, Nova Scotia, Canada: implications for the evolution of the Iapetus and 2280 Rheic Oceans. Tectonophysics 379, 199–209. https://doi.org/10.1016/j.tecto.2003.11.006 2281 Han, Z., Yang, Z., Tong, Y., Jing, X., 2015. New paleomagnetic results from Late Ordovician rocks of the 2282 Yangtze Block, South China, and their paleogeographic implications. J. Geophys. Res. [Solid 2283 Earth] 120, 4759–4772. 2284 Handke, M.J., Tucker, R.D., Ashwal, L.D., 1999. Neoproterozoic continental arc magmatism in west- 2285 central Madagascar. Geology 27, 351–354. 2.3.CO;2">https://doi.org/10.1130/0091- 2286 7613(1999)027<0351:NCAMIW>2.3.CO;2 2287 Hargraves, R.B., Dawson, E.M., Houten, F.B., 1987. Palaeomagnetism and age of mid-Palaeozoic ring 2288 complexes in Niger, , and tectonic implications. Geophys. J. Int. 90, 705–729. 2289 https://doi.org/10.1111/j.1365-246X.1987.tb00750.x 2290 Harlan, S.S., Geissman, J.W., Snee, L.W., 2008. Paleomagnetism of Proterozoic mafic dikes from the 2291 Tobacco Root Mountains, southwest Montana. Precambrian Res. 163, 239–264. 2292 https://doi.org/10.1016/j.precamres.2007.12.002 2293 Hartz, E.H., Torsvik, T.H., 2002. Baltica upside down: A new plate tectonic model for Rodinia and the 2294 Iapetus Ocean. Geology 30, 255–258. 2.0.CO;2">https://doi.org/10.1130/0091- 2295 7613(2002)030<0255:BUDANP>2.0.CO;2 2296 He, T., Zhu, M., Mills, B.J.W., Wynn, P.M., Zhuravlev, A.Y., Tostevin, R., Pogge von Strandmann, 2297 P.A.E., Yang, A., Poulton, S.W., Shields, G.A., 2019. Possible links between extreme oxygen 2298 perturbations and the Cambrian radiation of . Nat. Geosci. 12, 468–474. 2299 https://doi.org/10.1038/s41561-019-0357-z 2300 Heatherington, A.L., Mueller, P.A., Nutman, A.P., 1996. Neoproterozoic magmatism in the Suwannee 2301 terrane: Implications for terrane correlation. Geological Society of America Special Papers 304, 2302 257–268. 2303 Henry, B., Liégeois, J.P., Nouar, O., Derder, M.E.M., Bayou, B., Bruguier, O., Ouabadi, A., Belhai, D., 2304 Amenna, M., Hemmi, A., Ayache, M., 2009. Repeated granitoid intrusions during the 2305 Neoproterozoic along the western boundary of the , Eastern Hoggar, Tuareg 2306 shield, Algeria: An AMS and U–Pb zircon age study. Tectonophysics 474, 417–434. 2307 https://doi.org/10.1016/j.tecto.2009.04.022 2308 Heron, P.J., Murphy, J.B., Nance, D.R., Pysklywec, R.N., 2020. ’s mantle signature: the quest for 2309 supercontinent identification. Geological Society, London, Special Publications 503. 2310 https://doi.org/10.1144/SP503-2020-7 2311 Hibbard, J.P., van Staal, C.R., Miller, B.V., 2007. Links among Carolinia, Avalonia, and Ganderia in the 2312 Appalachian peri-Gondwanan realm. Geological Society of America Special Papers 433, 291– 2313 311. 2314 Hochmuth, K., Gohl, K., Uenzelmann-Neben, G., 2015. Playing jigsaw with large igneous provinces—A 2315 plate tectonic reconstruction of Ontong Java Nui, West Pacific. Geochem. Geophys. Geosyst. 16, 2316 3789–3807. 2317 Hodych, J.P., 1989. Limestones of western Newfoundland that magnetized before Devonian folding but 2318 after Middle Ordovician lithification. Geophys. Res. Lett. 16, 93–96. 2319 https://doi.org/10.1029/GL016i001p00093 2320 Hodych, J.P., Buchan, K.L., 1998. Palaeomagnetism of the ca. 440 Ma Cape St Mary’s sills of the Avalon 2321 Peninsula of Newfoundland: implications for Iapetus Ocean closure. Geophys. J. Int. 135, 155– 2322 164. https://doi.org/10.1046/j.1365-246X.1998.00263.x 2323 Hodych, J.P., Cox, R.A., Košler, J., 2004. An equatorial Laurentia at 550 Ma confirmed by Grenvillian 2324 inherited zircons dated by LAM ICP-MS in the Skinner Cove volcanics of western 2325 Newfoundland: implications for inertial interchange true polar wander. Precambrian Res. 129, 2326 93–113. https://doi.org/10.1016/j.precamres.2003.10.012 2327 Hoffman, P.F., 1991. Did the breakout of Laurentia turn Gondwanaland inside-out? Science 252, 1409– 2328 1412. https://doi.org/10.1126/science.252.5011.1409 2329 Hoffman, P.F., Kaufman, A.J., Halverson, G.P., Schrag, D.P., 1998. A Neoproterozoic snowball earth. 2330 Science 281, 1342–1346. https://doi.org/10.1126/science.281.5381.1342 2331 Hoffman, P.F., Schrag, D.P., 2002. The snowball Earth hypothesis: testing the limits of global change. 2332 Terra Nova 14, 129–155. https://doi.org/10.1046/j.1365-3121.2002.00408.x 2333 Hofmann, M., Linnemann, U., Rai, V., Becker, S., Gärtner, A., Sagawe, A., 2011. The India and South 2334 China cratons at the margin of Rodinia—Synchronous Neoproterozoic magmatism revealed by 2335 LA-ICP-MS zircon analyses. Lithos 123, 176–187. 2336 Hopper, E., Fischer, K.M., Wagner, L.S., Hawman, R.B., 2017. Reconstructing the end of the 2337 Appalachian orogeny. Geology 45, 15–18. https://doi.org/10.1130/G38453.1 2338 Hounslow, M.W., Domeier, M., Biggin, A.J., 2018. Subduction flux modulates the geomagnetic polarity 2339 reversal rate. Tectonophysics 742–743, 34–49. https://doi.org/10.1016/j.tecto.2018.05.018 2340 Hu, P.-Y., Zhai, Q.-G., Wang, J., Tang, Y., Wang, H.-T., Hou, K.-J., 2018. Precambrian origin of the 2341 North Lhasa terrane, : Constraint from early Cryogenian back-arc magmatism. 2342 Precambrian Res. 313, 51–67. https://doi.org/10.1016/j.precamres.2018.05.014 2343 Hu, X., Garzanti, E., Wang, J., Huang, W., An, W., Webb, A., 2016. The timing of India-Asia collision 2344 onset--Facts, theories, controversies. Earth-Sci. Rev. 160, 264–299. 2345 Huang, B., He, Z., Zong, K., Zhang, Z., 2014. Zircon U–Pb and Hf isotopic study of Neoproterozoic 2346 granitic gneisses from the Alatage area, : constraints on the Precambrian crustal 2347 evolution in the Central Tianshan Block. Chin. Sci. Bull. 59, 100–112. 2348 https://doi.org/10.1007/s11434-013-0010-y 2349 Huang, B., Xu, B., Zhang, C., Li, Y., Zhu, R., 2005. Paleomagnetism of the Baiyisi volcanic rocks (ca. 2350 740Ma) of Tarim, Northwest China: A of Neoproterozoic Western 2351 Australia? Precambrian Res. 142, 83–92. https://doi.org/10.1016/j.precamres.2005.09.006 2352 Huang, B., Yang, Z., Otofuji, Y.-I., Zhu, R., 1999. Early Paleozoic paleomagnetic poles from the western 2353 part of the North China Block and their implications. Tectonophysics 308, 377–402. 2354 https://doi.org/10.1016/S0040-1951(99)00098-0 2355 Huang, B.C., Otofuji, Y., Yang, Z.Y., 1999. Paleomagnetic constraints on the tectonic relationship 2356 between the Alashan/Hexi Corridor Terrane and the North China Block. Geophys. Res. Lett. 26, 2357 787–790. https://doi.org/10.1029/1999GL900097 2358 Huang, H., Cawood, P.A., Hou, M., Xiong, F., Ni, S., 2019. Provenance of latest Mesoproterozoic to 2359 early Neoproterozoic (meta)-sedimentary rocks and implications for paleographic reconstruction 2360 of the Yili Block. Gondwana Res. 2361 Huang, K., Opdyke, N.D., Zhu, R., 2000. Further paleomagnetic results from the Silurian of the Yangtze 2362 Block and their implications. Earth Planet. Sci. Lett. 175, 191–202. 2363 https://doi.org/10.1016/S0012-821X(99)00302-7 2364 Huang, Z., Long, X., Kröner, A., Yuan, C., Wang, Y., Chen, B., Zhang, Y., 2015. Neoproterozoic granitic 2365 gneisses in the Chinese Central Tianshan Block: Implications for tectonic affinity and 2366 Precambrian crustal evolution. Precambrian Res. 269, 73–89. 2367 https://doi.org/10.1016/j.precamres.2015.08.005 2368 Iosifidi, A.G., Khramov, A.N., Rodionov, V.P., Pisarevskii, S.A., Popov, V.V., 1999. Geomagnetic 2369 reversals in the Early Paleozoic: 2. A nonsynchronous record of Middle Ordovician reversals in 2370 the Berezovskaya section, southern Siberian Platform. Izv. Phys. Solid Earth 35, 24–32. 2371 Jackson, M., Van der Voo, R., 1985. A Lower Ordovician paleomagnetic pole from the Oneota dolomite, 2372 Upper Mississippi River Valley. J. Geophys. Res., Geol. Soc. Am. 90, 10449. 2373 https://doi.org/10.1029/JB090iB12p10449 2374 Jacobs, J., Elburg, M., Läufer, A., Kleinhanns, I.C., Henjes-Kunst, F., Estrada, S., Ruppel, A.S., 2375 Damaske, D., Montero, P., Bea, F., 2015. Two distinct Late Mesoproterozoic/Early 2376 Neoproterozoic basement provinces in central/eastern Dronning Maud Land, East Antarctica: The 2377 missing link, 15–21°E. Precambrian Res. 265, 249–272. 2378 https://doi.org/10.1016/j.precamres.2015.05.003 2379 Jacobs, J., Opås, B., Elburg, M.A., Läufer, A., Estrada, S., Ksienzyk, A.K., Damaske, D., Hofmann, M., 2380 2017. Cryptic sub-ice geology revealed by a U-Pb zircon study of glacial till in Dronning Maud 2381 Land, East Antarctica. Precambrian Res. 294, 1–14. 2382 https://doi.org/10.1016/j.precamres.2017.03.012 2383 Jarrard, R.D., 2003. Subduction fluxes of water, carbon dioxide, chlorine, and potassium. Geochem. 2384 Geophys. Geosyst., Geophys. Monogr. Ser. 4. https://doi.org/10.1029/2002GC000392 2385 Jeleńska, M., Bakhmutov, V., Konstantinienko, L., 2005. Paleomagnetic and rock magnetic data from the 2386 Silurian succession of the Dniester basin, Ukraine. Phys. Earth Planet. Inter. 149, 307–320. 2387 https://doi.org/10.1016/j.pepi.2004.10.005 2388 Jeleńska, M., Kądziałko-Hofmokl, M., Bakhmutov, V., Poliachenko, I., Ziółkowski, P., 2015. 2389 Palaeomagnetic and rock magnetic study of Lower Devonian sediments from Podolia, SW 2390 Ukraine: remagnetization problems. Geophys. J. Int. 200, 557–573. 2391 https://doi.org/10.1093/gji/ggu411 2392 Jian, P., Kröner, A., Jahn, B.-M., Windley, B.F., Shi, Y., Zhang, W., Zhang, F., Miao, L., Tomurhuu, D., 2393 Liu, D., 2014. Zircon dating of Neoproterozoic and Cambrian ophiolites in West Mongolia and 2394 implications for the timing of orogenic processes in the central part of the Central Asian Orogenic 2395 Belt. Earth-Sci. Rev. 133, 62–93. https://doi.org/10.1016/j.earscirev.2014.02.006 2396 Jiang, G., Sohl, L.E., Christie-Blick, N., 2003. Neoproterozoic stratigraphic comparison of the Lesser 2397 Himalaya (India) and Yangtze block (south China): Paleogeographic implications. Geology 31, 2398 917–920. https://doi.org/10.1130/G19790.1 2399 Jing, X.-Q., Yang, Z., Tong, Y., Han, Z., 2015. A revised paleomagnetic pole from the mid- 2400 Neoproterozoic Liantuo Formation in the Yangtze block and its paleogeographic implications. 2401 Precambrian Res. 268, 194–211. https://doi.org/10.1016/j.precamres.2015.07.007 2402 Johnson, P.R., Andresen, A., Collins, A.S., Fowler, A.R., Fritz, H., Ghebreab, W., Kusky, T., Stern, R.J., 2403 2011. Late Cryogenian--Ediacaran history of the Arabian--Nubian Shield: a review of 2404 depositional, plutonic, structural, and tectonic events in the closing stages of the northern East 2405 African Orogen. J. Afr. Earth. Sci. 61, 167–232. 2406 Johnson, R.J.E., Van Der Voo, R., 1990. Pre-folding magnetization reconfirmed for the Late Ordovician- 2407 Early Silurian Dunn Point volcanics, Nova Scotia. Tectonophysics 178, 193–205. 2408 https://doi.org/10.1016/0040-1951(90)90146-Y 2409 Johnson, R.J.E., Van der Voo, R., 1985. Middle Cambrian paleomagnetism of the Avalon Terrane in 2410 Cape Breton Island, Nova Scotia. Tectonics 4, 629–651. 2411 https://doi.org/10.1029/TC004i007p00629 2412 Just, J., Schulz, B., de Wall, H., Jourdan, F., Pandit, M.K., 2011. Monazite CHIME/EPMA dating of 2413 Erinpura granitoid deformation: Implications for Neoproterozoic tectono-thermal evolution of 2414 NW India. Gondwana Res. 19, 402–412. https://doi.org/10.1016/j.gr.2010.08.002 2415 Kamo, S.L., Gower, C.F., 1994. Note: U-Pb baddeleyite dating clarifies age of characteristic 2416 paleomagnetic remanence of Long Range dykes, southeastern Labrador. 2417 Karlsen, K.S., Conrad, C.P., Magni, V., 2019. Deep Water Cycling and Change Since the 2418 Breakup of Pangea. Geochem. Geophys. Geosyst. 20, 2919–2935. 2419 https://doi.org/10.1029/2019GC008232 2420 Keppie, J.D., Davis, D.W., Krogh, T.E., 1998. U-Pb geochronological constraints on Precambrian 2421 stratified units in the Avalon Composite Terrane of Nova Scotia, Canada: tectonic implications. 2422 Can. J. Earth Sci. 35, 222–236. https://doi.org/10.1139/e97-109 2423 Keppie, J.D., Dostal, J., 1998. Birth of the Avalon arc in Nova Scotia, Canada: geochemical evidence for 2424 700--630 Ma back-arc rift volcanism off Gondwana. Geol. Mag. 135, 171–181. 2425 Khramov, A.N., Iosifidi, A.G., 2009. Paleomagnetism of the Lower Ordovician and Cambrian 2426 sedimentary rocks in the section of the Narva river right bank: For the construction of the Baltic 2427 kinematic model in the Early Paleozoic. Izv. Phys. Solid Earth 45, 465–481. 2428 Kirschvink, J.L., Ripperdan, R.L., Evans, D.A., 1997. Evidence for a Large-Scale Reorganization of Early 2429 Cambrian Continental Masses by Inertial Interchange True Polar Wander. Science 277, 541–545. 2430 https://doi.org/10.1126/science.277.5325.541 2431 Knesel, K.M., Cohen, B.E., Vasconcelos, P.M., Thiede, D.S., 2008. Rapid change in drift of the 2432 records collision with . Nature 454, 754–757. 2433 https://doi.org/10.1038/nature07138 2434 Konstantinovskaya, E.A., 2002. The mechanism of continental crust accretion: an example of Western 2435 Kamchatka. Geotectonics/Geotektonika 36, 393–411. 2436 Krogh, T.E., Strong, D.F., O’Brien, S.J., Papezik, V.S., 1988. Precise U–Pb zircon dates from the Avalon 2437 Terrane in Newfoundland. Can. J. Earth Sci. 25, 442–453. https://doi.org/10.1139/e88-045 2438 Kröner, A., Alexeiev, D.V., Hegner, E., Rojas-Agramonte, Y., Corsini, M., Chao, Y., Wong, J., Windley, 2439 B.F., Liu, D., Tretyakov, A.A., 2012. Zircon and muscovite ages, geochemistry, and Nd–Hf 2440 isotopes for the Aktyuz metamorphic terrane: Evidence for an Early Ordovician collisional belt in 2441 the northern Tianshan of Kyrgyzstan. Gondwana Res. 21, 901–927. 2442 https://doi.org/10.1016/j.gr.2011.05.010 2443 Kroner, A., Windley, B.F., Badarch, G., Tomurtogoo, O., Hegner, E., Jahn, B.M., Gruschka, S., Khain, 2444 E.V., Demoux, A., Wingate, M.T.D., 2007. Accretionary growth and crust formation in the 2445 Central Asian Orogenic Belt and comparison with the Arabian-Nubian shield. Memoirs- 2446 Geological Society of America 200, 181. 2447 Lapique, F., Bertrand, J.M., Meriem, D., 1986. A major Pan-African crustal decoupling zone in the 2448 Timgaouine area (Western Hoggar, Algeria). Journal of African Earth Sciences (1983). 2449 https://doi.org/10.1016/0899-5362(86)90028-x 2450 Lenton, T.M., Dahl, T.W., Daines, S.J., Mills, B.J.W., Ozaki, K., Saltzman, M.R., Porada, P., 2016. 2451 Earliest land plants created modern levels of atmospheric oxygen. Proc. Natl. Acad. Sci. U. S. A. 2452 113, 9704–9709. https://doi.org/10.1073/pnas.1604787113 2453 Levashova, N.M., Bazhenov, M.L., Meert, J.G., Kuznetsov, N.B., Golovanova, I.V., Danukalov, K.N., 2454 Fedorova, N.M., 2013. Paleogeography of Baltica in the Ediacaran: Paleomagnetic and 2455 geochronological data from the clastic Zigan Formation, South Urals. Precambrian Res. 236, 16– 2456 30. https://doi.org/10.1016/j.precamres.2013.06.006 2457 Li, M., McNamara, A.K., 2013. The difficulty for subducted oceanic crust to accumulate at the Earth’s 2458 core-mantle boundary. J. Geophys. Res. [Solid Earth] 118, 1807–1816. 2459 Li, Yianping, McWillams, M., Sharps, R., Cox, A., Li, Yongan, Li, Q., Gao, Z., Zhang, Z., Zhai, Y., 2460 1990. A Devonian paleomagnetic pole from red beds of the Tarim Block, China. J. Geophys. 2461 Res., Geodyn. Ser. 95, 19185. https://doi.org/10.1029/JB095iB12p19185 2462 Li, Z., Ding, L., Lippert, P.C., Song, P., Yue, Y., van Hinsbergen, D.J.J., 2016. Paleomagnetic constraints 2463 on the Mesozoic drift of the Lhasa terrane (Tibet) from Gondwana to Eurasia. Geology 44, 727– 2464 730. https://doi.org/10.1130/G38030.1 2465 Li, Z., Ding, L., Song, P., Fu, J., Yue, Y., 2017. Paleomagnetic constraints on the paleolatitude of the 2466 Lhasa block during the Early Cretaceous: implications for the onset of India--Asia collision and 2467 latitudinal shortening estimates across Tibet and stable Asia. Gondwana Res. 41, 352–372. 2468 Li, Z.X., Bogdanova, S.V., Collins, A.S., Davidson, A., De Waele, B., Ernst, R.E., Fitzsimons, I.C.W., 2469 Fuck, R.A., Gladkochub, D.P., Jacobs, J., Karlstrom, K.E., Lu, S., Natapov, L.M., Pease, V., 2470 Pisarevsky, S.A., Thrane, K., Vernikovsky, V., 2008. Assembly, configuration, and break-up 2471 history of Rodinia: A synthesis. Precambrian Res. 160, 179–210. 2472 https://doi.org/10.1016/j.precamres.2007.04.021 2473 Li, Z.-X., Evans, D.A.D., Halverson, G.P., 2013. Neoproterozoic glaciations in a revised global 2474 palaeogeography from the breakup of Rodinia to the assembly of Gondwanaland. Sediment. 2475 Geol. 294, 219–232. https://doi.org/10.1016/j.sedgeo.2013.05.016 2476 Li, Z.X., Evans, D.A.D., Zhang, S., 2004. A 90° spin on Rodinia: possible causal links between the 2477 Neoproterozoic supercontinent, superplume, true polar wander and low-latitude glaciation. Earth 2478 Planet. Sci. Lett. 220, 409–421. https://doi.org/10.1016/S0012-821X(04)00064-0 2479 Li, Z.X., Mitchell, R.N., Spencer, C.J., Ernst, R., Pisarevsky, S., Kirscher, U., Murphy, J.B., 2019. 2480 Decoding Earth’s rhythms: modulation of supercontinent cycles by longer episodes. 2481 Precambrian Res. 323, 1–5. 2482 Li, Z.-X., Zhang, L., Powell, C.M., 1995. South China in Rodinia: Part of the missing link between 2483 Australia–East Antarctica and Laurentia? Geology 23, 407–410. 2484 2.3.CO;2">https://doi.org/10.1130/0091-7613(1995)023<0407:SCIRPO>2.3.CO;2 2485 Li, Z.-X., Zhong, S., 2009. Supercontinent–superplume coupling, true polar wander and plume mobility: 2486 Plate dominance in whole-mantle tectonics. Phys. Earth Planet. Inter. 176, 143–156. 2487 https://doi.org/10.1016/j.pepi.2009.05.004 2488 Liégeois, J.P., Black, R., Navez, J., Latouche, L., 1994. Early and late Pan-African orogenies in the Aïr 2489 assembly of terranes (Tuareg shield, Niger). Precambrian Res. 67, 59–88. 2490 https://doi.org/10.1016/0301-9268(94)90005-1 2491 Liégeois, J.P., Latouche, L., Boughrara, M., Navez, J., Guiraud, M., 2003. The LATEA metacraton 2492 (Central Hoggar, Tuareg shield, Algeria): behaviour of an old passive margin during the Pan- 2493 African orogeny. J. Afr. Earth. Sci. 37, 161–190. https://doi.org/10.1016/j.jafrearsci.2003.05.004 2494 Liu, L., Gurnis, M., Seton, M., Saleeby, J., Müller, R.D., Jackson, J.M., 2010. The role of 2495 subduction in the Laramide orogeny. Nat. Geosci. 3, 353–357. https://doi.org/10.1038/ngeo829 2496 Liu, L., Spasojevic, S., Gurnis, M., 2008. Reconstructing subduction beneath North 2497 America back to the Late Cretaceous. Science 322, 934–938. 2498 https://doi.org/10.1126/science.1162921 2499 Liu, X., Zhao, Y., Chen, H., Song, B., 2017. New zircon U–Pb and Hf–Nd isotopic constraints on the 2500 timing of magmatism, sedimentation and metamorphism in the northern Prince Charles 2501 Mountains, East Antarctica. Precambrian Res. 299, 15–33. 2502 https://doi.org/10.1016/j.precamres.2017.07.012 2503 Lixin, B., Rixiang, Z., Harming, W., Bin, G., Jianjun, L., 1998. New Cambrian paleomagnetic pole for 2504 Yangtze Block. Sci. China Ser. D Earth Sci. 41, 66–71. https://doi.org/10.1007/BF02984514 2505 Llanos, M.P.I., Tait, J.A., Popov, V., Abalmassova, A., 2005. Palaeomagnetic data from Ediacaran 2506 (Vendian) sediments of the region, NW Russia: An alternative apparent polar 2507 wander path of Baltica for the Late Proterozoic–Early Palaeozoic. Earth Planet. Sci. Lett. 240, 2508 732–747. https://doi.org/10.1016/j.epsl.2005.09.063 2509 Loucks, V., Elmore, D.R., 1986. Absolute dating of dedolomitization and the origin of magnetization in 2510 the Cambrian Morgan Creek Limestone, central Texas. GSA Bulletin 97, 486–496. 2511 2.0.CO;2">https://doi.org/10.1130/0016-7606(1986)97<486:ADODAT>2.0.CO;2 2512 Lubnina, N.V., Iosifidi, A.G., Khramov, A.N., Popov, V.V., Lewandowski, M., 2007. Paleomagnetism of 2513 the Silurian and Devoni-an sedimentary formations of Podolia, Ukraine, in: Khramov, A.N. (Ed.), 2514 Paleomagnetism of the Northern Eurasia Sedimentary Basins (IN RUSSIAN). pp. 105–125. 2515 Lubnina, N.V., Pisarevsky, S.A., Puchkov, V.N., Kozlov, V.I., Sergeeva, N.D., 2014. New paleomagnetic 2516 data from Late Neoproterozoic sedimentary successions in Southern Urals, Russia: implications 2517 for the Late Neoproterozoic paleogeography of the Iapetan realm. Int. J. Earth Sci. 103, 1317– 2518 1334. https://doi.org/10.1007/s00531-014-1013-x 2519 Macdonald, F.A., Prave, A.R., Petterson, R., Smith, E.F., Pruss, S.B., Oates, K., Waechter, F., Trotzuk, 2520 D., Fallick, A.E., 2013. The Laurentian record of Neoproterozoic glaciation, tectonism, and 2521 eukaryotic evolution in Death Valley, . GSA Bulletin 125, 1203–1223. 2522 https://doi.org/10.1130/B30789.1 2523 Martin, M.W., Grazhdankin, D.V., Bowring, S.A., Evans, D.A., Fedonkin, M.A., Kirschvink, J.L., 2000. 2524 Age of Neoproterozoic bilatarian body and trace fossils, , Russia: implications for 2525 metazoan evolution. Science 288, 841–845. https://doi.org/10.1126/science.288.5467.841 2526 Matthews, K.J., Maloney, K.T., Zahirovic, S., Williams, S.E., Seton, M., Müller, R.D., 2016. Global plate 2527 boundary evolution and kinematics since the late Paleozoic. Glob. Planet. Change 146, 226–250. 2528 https://doi.org/10.1016/j.gloplacha.2016.10.002 2529 Matthews, K.J., Williams, S.E., Whittaker, J.M., Müller, R.D., Seton, M., Clarke, G.L., 2015. Geologic 2530 and kinematic constraints on Late Cretaceous to mid plate boundaries in the southwest 2531 Pacific. Earth-Sci. Rev. 140, 72–107. https://doi.org/10.1016/j.earscirev.2014.10.008 2532 McCabe, C., Channell, J.E.T., 1990. Paleomagnetic results from volcanic rocks of the Shelve Inlier, 2533 Wales: evidence for a wide Late Ordovician Iapetus Ocean in Britain. Earth Planet. Sci. Lett. 96, 2534 458–468. https://doi.org/10.1016/0012-821X(90)90020-X 2535 McCabe, C., Channell, J.E.T., Woodcock, N.H., 1992. Further paleomagnetic results from the Builth 2536 Wells Ordovician inlier, Wales. J. Geophys. Res. [Solid Earth] 97, 9357–9370. 2537 McCabe, C., Van der Voo, R., Wilkinson, B.H., Devaney, K., 1985. A Middle/Late Silurian 2538 paleomagnetic pole from limestone reefs of the Wabash Formation, Indiana, USA. J. Geophys. 2539 Res. [Solid Earth] 90, 2959–2965. 2540 McCausland, P.J.A., Hodych, J.P., 1998. Paleomagnetism of the 550 Ma Skinner Cove volcanics of 2541 western Newfoundland and the opening of the Iapetus Ocean. Earth Planet. Sci. Lett. 163, 15–29. 2542 https://doi.org/10.1016/S0012-821X(98)00171-X 2543 McCausland, P.J.A., Van der Voo, R., Hall, C.M., 2007. Circum-Iapetus paleogeography of the 2544 Precambrian–Cambrian transition with a new paleomagnetic constraint from Laurentia. 2545 Precambrian Res. 156, 125–152. https://doi.org/10.1016/j.precamres.2007.03.004 2546 McElhinny, M.W., Cowley, J.A., Edwards, D.J., 1978. Palaeomagnetism of some rocks from Peninsular 2547 India and Kashmir. Tectonophysics 50, 41–54. https://doi.org/10.1016/0040-1951(78)90198-1 2548 McGee, B., Halverson, G.P., Collins, A.S., 2012. Cryogenian rift-related magmatism and sedimentation: 2549 South-western Congo Craton, Namibia. J. Afr. Earth. Sci. 76, 34–49. 2550 https://doi.org/10.1016/j.jafrearsci.2012.09.003 2551 McKenzie, D.P., Parker, R.L., 1967. The North Pacific: an Example of Tectonics on a Sphere. Nature 2552 216, 1276–1280. https://doi.org/10.1038/2161276a0 2553 McKenzie, R.N., Hughes, N.C., Gill, B.C., Myrow, P.M., 2014. Plate tectonic influences on 2554 Neoproterozoic–early Paleozoic climate and animal evolution. Geology 42, 127–130. 2555 https://doi.org/10.1130/G34962.1 2556 McKerrow, W.S., Mac Niocaill, C., Ahlberg, P.E., Clayton, G., Cleal, C.J., Eagar, R.M.C., 2000. The 2557 Late Palaeozoic relations between Gondwana and Laurussia. Geological Society, London, Special 2558 Publications 179, 9–20. https://doi.org/10.1144/GSL.SP.2000.179.01.03 2559 McMenamin, M.A., McMenamin, M.A., McMenamin, D.L.S., 1990. The emergence of animals: the 2560 Cambrian breakthrough. Columbia University Press. 2561 Meert, J.G., 2014. Ediacaran–Early Ordovician paleomagnetism of Baltica: A review. Gondwana Res. 25, 2562 159–169. https://doi.org/10.1016/j.gr.2013.02.003 2563 Meert, J.G., 2003. A synopsis of events related to the assembly of eastern Gondwana. Tectonophysics 2564 362, 1–40. https://doi.org/10.1016/S0040-1951(02)00629-7 2565 Meert, J.G., 2002. Paleomagnetic Evidence for a Paleo-Mesoproterozoic Supercontinent Columbia. 2566 Gondwana Res. 5, 207–215. https://doi.org/10.1016/S1342-937X(05)70904-7 2567 Meert, J.G., Lieberman, B.S., 2008. The Neoproterozoic assembly of Gondwana and its relationship to 2568 the Ediacaran–Cambrian radiation. Gondwana Res. 14, 5–21. 2569 https://doi.org/10.1016/j.gr.2007.06.007 2570 Meert, J.G., Pandit, M.K., Kamenov, G.D., 2013. Further geochronological and paleomagnetic constraints 2571 on Malani (and pre-Malani) magmatism in NW India. Tectonophysics 608, 1254–1267. 2572 https://doi.org/10.1016/j.tecto.2013.06.019 2573 Meert, J.G., Pandit, M.K., Pradhan, V.R., Banks, J., Sirianni, R., Stroud, M., Newstead, B., Gifford, J., 2574 2010. Precambrian crustal evolution of Peninsular India: A 3.0 billion odyssey. J. Asian 2575 Earth Sci. 39, 483–515. https://doi.org/10.1016/j.jseaes.2010.04.026 2576 Meert, J.G., van der Voo, R., Ayub, S., 1995. Paleomagnetic investigation of the Neoproterozoic Gagwe 2577 lavas and Mbozi complex, Tanzania and the assembly of Gondwana. Precambrian Res. 74, 225– 2578 244. https://doi.org/10.1016/0301-9268(95)00012-T 2579 Meffre, S., Direen, N.G., Crawford, A.J., Kamenetsky, V., 2004. Mafic volcanic rocks on King Island, 2580 Tasmania: evidence for 579 Ma break-up in east Gondwana. Precambrian Res. 135, 177–191. 2581 Merdith, A.S., Atkins, S.E., Tetley, M.G., 2019a. Tectonic Controls on Carbon and Storage 2582 in Subducted Upper Oceanic Lithosphere for the Past 320 Ma. Front Earth Sci. Chin. 7, 332. 2583 https://doi.org/10.3389/feart.2019.00332 2584 Merdith, A.S., Collins, A.S., Williams, S.E., Pisarevsky, S., Foden, J.D., Archibald, D.B., Blades, M.L., 2585 Alessio, B.L., Armistead, S., Plavsa, D., Clark, C., Müller, R.D., 2017a. A full-plate global 2586 reconstruction of the Neoproterozoic. Gondwana Res. 50, 84–134. 2587 https://doi.org/10.1016/j.gr.2017.04.001 2588 Merdith, A.S., Williams, S.E., Brune, S., Collins, A.S., Müller, R.D., 2019b. Rift and plate boundary 2589 evolution across two supercontinent cycles. Glob. Planet. Change 173, 1–14. 2590 https://doi.org/10.1016/j.gloplacha.2018.11.006 2591 Merdith, A.S., Williams, S.E., Müller, R.D., Collins, A.S., 2017b. Kinematic constraints on the Rodinia to 2592 Gondwana transition. Precambrian Res. 299, 132–150. 2593 https://doi.org/10.1016/j.precamres.2017.07.013 2594 Metcalfe, I., 2013. Gondwana dispersion and Asian accretion: Tectonic and palaeogeographic evolution 2595 of eastern Tethys. J. Asian Earth Sci. 66, 1–33. https://doi.org/10.1016/j.jseaes.2012.12.020 2596 Metcalfe, I., 2011. Palaeozoic–Mesozoic history of SE Asia. Geological Society, London, Special 2597 Publications 355, 7–35. https://doi.org/10.1144/SP355.2 2598 Metelkin, D.V., Vernikovsky, V.A., Kazansky, A.Y., 2012. Tectonic evolution of the Siberian 2599 from the Neoproterozoic to the Late Mesozoic: paleomagnetic record and 2600 reconstructions. Russ. Geol. Geophys. 53, 675–688. https://doi.org/10.1016/j.rgg.2012.05.006 2601 Miller, J.D., Kent, D.V., 1988. Paleomagnetism of the Silurian-Devonian Andreas redbeds: Evidence for 2602 an Early Devonian supercontinent? Geology 16, 195–198. 2603 2.3.CO;2">https://doi.org/10.1130/0091-7613(1988)016<0195:POTSDA>2.3.CO;2 2604 Mills, B., Watson, A.J., Goldblatt, C., Boyle, R., Lenton, T.M., 2011. Timing of Neoproterozoic 2605 glaciations linked to transport-limited global . Nat. Geosci. 4, 861–864. 2606 https://doi.org/10.1038/ngeo1305 2607 Mills, B.J.W., Krause, A.J., Scotese, C.R., Hill, D.J., Shields, G.A., Lenton, T.M., 2019. Modelling the 2608 long-term carbon cycle, atmospheric CO2, and Earth surface temperature from late 2609 Neoproterozoic to present day. Gondwana Res. 67, 172–186. 2610 https://doi.org/10.1016/j.gr.2018.12.001 2611 Millward, D., Evans, J.A., 2003. U–Pb chronology and duration of late Ordovician magmatism in the 2612 English District. J. Geol. Soc. London 160, 773–781. https://doi.org/10.1144/0016-764902- 2613 160 2614 Minh, P., Hieu, P.T., Thuy, N.T.B., Dung, L.T., Kawaguchi, K., Dung, P.T., 2020. Neoproterozoic 2615 granitoids from the Phan Si Pan Zone, NW Vietnam: geochemistry and geochronology 2616 constraints on reconstructing South China--India Palaeogeography. Int. Geol. Rev. 1–16. 2617 Minibaev, R.A. and Sulutdinov, R.M., 1991. Paleomagnitniye kharakteristiki siluriyskikh obrazovanii 2618 sakmarskoy zoni Yuzhnogo Urala. Bashkirian Natl.Cttee.Ural.Branch Acad.Sci.USSR (Ufa) 30. 2619 Mitchell, R.N., Kilian, T.M., Evans, D.A.D., 2012. Supercontinent cycles and the calculation of absolute 2620 palaeolongitude in deep time. Nature 482, 208–211. https://doi.org/10.1038/nature10800 2621 Moores, E.M., 2002. Pre–1 Ga (pre-Rodinian) ophiolites: Their tectonic and environmental implications. 2622 GSA Bulletin 114, 80–95. 2.0.CO;2">https://doi.org/10.1130/0016- 2623 7606(2002)114<0080:PGPROT>2.0.CO;2 2624 Moores, E.M., 1991. Southwest U.S.-East Antarctic (SWEAT) connection: A hypothesis. Geology 19, 2625 425–428. 2.3.CO;2">https://doi.org/10.1130/0091-7613(1991)019<0425:SUSEAS>2.3.CO;2 2626 Morgan, W.J., 1968. Rises, trenches, great faults, and crustal blocks. J. Geophys. Res. 73, 1959–1982. 2627 https://doi.org/10.1029/JB073i006p01959 2628 Morrison, J., Ellwood, B.B., 1986. Paleomagnetism of Silurian-Ordovician sediments from the Valley 2629 and Ridge province, northwest Georgia. Geophys. Res. Lett. 13, 189–192. 2630 Mulder, J.A., Berry, R.F., Halpin, J.A., Meffre, S., Everard, J.L., 2018a. Depositional age and correlation 2631 of the Oonah Formation: refining the timing of Neoproterozoic basin formation in Tasmania. 2632 Aust. J. Earth Sci. 65, 391–407. https://doi.org/10.1080/08120099.2018.1426629 2633 Mulder, J.A., Everard, J.L., Cumming, G., Meffre, S., Bottrill, R.S., Merdith, A.S., Halpin, J.A., McNeill, 2634 A.W., Cawood, P.A., 2020. Neoproterozoic opening of the Pacific Ocean recorded by multi-stage 2635 rifting in Tasmania, Australia. Earth-Sci. Rev. 201, 103041. 2636 https://doi.org/10.1016/j.earscirev.2019.103041 2637 Mulder, J.A., Halpin, J.A., Daczko, N.R., 2015. Mesoproterozoic Tasmania: Witness to the East 2638 Antarctica–Laurentia connection within Nuna. Geology 43, 759–762. 2639 https://doi.org/10.1130/G36850.1 2640 Mulder, J.A., Halpin, J.A., Daczko, N.R., Orth, K., Meffre, S., Thompson, J.M., Morrissey, L.J., 2019. A 2641 Multiproxy provenance approach to uncovering the assembly of East Gondwana in Antarctica. 2642 Geology 47, 645–649. https://doi.org/10.1130/G45952.1 2643 Mulder, J.A., Karlstrom, K.E., Halpin, J.A., Merdith, A.S., Spencer, C.J., Berry, R.F., McDonald, B., 2644 2018b. Rodinian devil in disguise: Correlation of 1.25–1.10 Ga strata between Tasmania and 2645 Grand Canyon. Geology 46, 991–994. https://doi.org/10.1130/G45225.1 2646 Müller, R.D., Cannon, J., Qin, X., Watson, R.J., Gurnis, M., Williams, S., Pfaffelmoser, T., Seton, M., 2647 Russell, S.H.J., Zahirovic, S., 2018. GPlates: Building a Virtual Earth Through Deep Time. 2648 Geochem. Geophys. Geosyst. 19, 2243–2261. https://doi.org/10.1029/2018GC007584 2649 Müller, R.D., Royer, J.-Y., Lawver, L.A., 1993. Revised plate motions relative to the hotspots from 2650 combined Atlantic and Indian Ocean hotspot tracks. Geology 21, 275–278. 2651 2.3.CO;2">https://doi.org/10.1130/0091-7613(1993)021<0275:RPMRTT>2.3.CO;2 2652 Müller, R.D., Sdrolias, M., Gaina, C., Steinberger, B., Heine, C., 2008. Long-term sea-level fluctuations 2653 driven by ocean basin dynamics. Science 319, 1357–1362. 2654 https://doi.org/10.1126/science.1151540 2655 Müller, R.D., Seton, M., Zahirovic, S., Williams, S.E., Matthews, K.J., Wright, N.M., Shephard, G.E., 2656 Maloney, K.T., Barnett-Moore, N., Hosseinpour, M., Bower, D.J., Cannon, J., 2016. Ocean Basin 2657 Evolution and Global-Scale Plate Reorganization Events Since Pangea Breakup. Annu. Rev. 2658 Earth Planet. Sci. 44, 107–138. https://doi.org/10.1146/annurev-earth-060115-012211 2659 Müller, R.D., Zahirovic, S., Williams, S.E., Cannon, J., Seton, M., Bower, D.J., Tetley, M., Heine, C., Le 2660 Breton, E., Liu, S., Russell, S.H.J., Yang, T., Leonard, J., Gurnis, M., 2019. A global plate model 2661 including lithospheric deformation along major rifts and orogens since the Triassic. Tectonics. 2662 https://doi.org/10.1029/2018TC005462 2663 Murphy, J.B., 2002. Geochemistry of the Neoproterozoic metasedimentary Gamble Brook Formation, 2664 Avalon terrane, Nova Scotia: evidence for a rifted-arc environment along the west Gondwanan 2665 margin of Rodinia. J. Geol. 110, 407–419. 2666 Murphy, J.B., Keppie, J.D., Dostal, J., Nance, R.D., 1999. Neoproterozoic-early Paleozoic evolution 2667 ofAvalonia. Laurentia-Gondwana Connections Before Pangea 336, 253. 2668 Murphy, J.B., Nance, R.D., 1989. Model for the evolution of the Avalonian-Cadomian belt. Geology 17, 2669 735–738. 2.3.CO;2">https://doi.org/10.1130/0091-7613(1989)017<0735:MFTEOT>2.3.CO;2 2670 Murphy, J.B., Pisarevsky, S.A., Nance, R.D., Keppie, J.D., 2004. Neoproterozoic—Early Paleozoic 2671 evolution of peri-Gondwanan terranes: implications for Laurentia-Gondwana connections. Int. J. 2672 Earth Sci. 93, 659–682. https://doi.org/10.1007/s00531-004-0412-9 2673 Murphy, J.B., Strachan, R.A., Nance, R.D., Parker, K.D., Fowler, M.B., 2000. Proto-Avalonia: A 1.2–1.0 2674 Ga tectonothermal event and constraints for the evolution of Rodinia. Geology 28, 1071–1074. 2675 2.0.CO;2">https://doi.org/10.1130/0091-7613(2000)28<1071:PAGTEA>2.0.CO;2 2676 Murphy, J.B., Waldron, J.W.F., Kontak, D.J., Pe-Piper, G., Piper, D.J.W., 2011. Minas Fault Zone: Late 2677 Paleozoic history of an intra-continental orogenic transform fault in the Canadian Appalachians. 2678 J. Struct. Geol. 33, 312–328. https://doi.org/10.1016/j.jsg.2010.11.012 2679 Murthy, G., Gower, C., Tubrett, M., Pätzold, R., 1992. Paleomagnetism of Eocambrian Long Range 2680 dykes and Double Mer Formation from Labrador, Canada. Can. J. Earth Sci. 29, 1224–1234. 2681 https://doi.org/10.1139/e92-098 2682 Nance, R.D., Linnemann, U., 2008. The Rheic Ocean: origin, evolution, and significance. GSA Today 18, 2683 4–12. 2684 Nance, R.D., Murphy, J.B., Strachan, R.A., Keppie, J.D., Gutiérrez-Alonso, G., Fernández-Suárez, J., 2685 Quesada, C., Linnemann, U., D’lemos, R., Pisarevsky, S.A., 2008. Neoproterozoic-early 2686 Palaeozoic tectonostratigraphy and palaeogeography of the peri-Gondwanan terranes: Amazonian 2687 v. West African connections. Geological Society, London, Special Publications 297, 345–383. 2688 Nick, K.E., Elmore, R.D., 1990. Paleomagnetism of the Cambrian Royer Dolomite and 2689 Collings Ranch Conglomerate, southern Oklahoma: An early Paleozoic magnetization and 2690 nonpervasive remagnetization by weathering. Geol. Soc. Am. Bull. 102, 1517–1525. 2691 Niu, J., Li, Z.-X., Zhu, W., 2016. Palaeomagnetism and geochronology of mid-Neoproterozoic Yanbian 2692 dykes, South China: implications for a c. 820–800 Ma true polar wander event and the 2693 reconstruction of Rodinia. Geological Society, London, Special Publications 424, 191–121. 2694 https://doi.org/http: //doi.org/10.1144/SP424.11 2695 Noel, J.R., Spariosu, D.J., Dallmeyer, R.D., 1988. Paleomagnetism and 40Ar/39Ar ages from the 2696 Carolina slate belt, Albemarle, North Carolina: Implications for terrane amalgamation with North 2697 America. Geology 16, 64–68. 2698 O’Brien, S.J., Dunning, G.R., Dube, B., O’Driscoll, C.F., Sparkes, B., Israel, S., Ketchum, J., 2001. New 2699 insights into the Neoproterozoic geology of the central Avalon Peninsula (parts of NTS map areas 2700 1N/6, 1N/7 and 1N/3), eastern Newfoundland. Current Research, Government of Newfoundland 2701 and Labrador, Department of Mines and Energy, Geological Survey 2001, 169–189. 2702 O’Neill, C., Müller, D., Steinberger, B., 2005. On the uncertainties in hot spot reconstructions and the 2703 significance of moving hot spot reference frames. Geochem. Geophys. Geosyst., Spec. Pap. Geol. 2704 Soc. Am. 6. https://doi.org/10.1029/2004GC000784 2705 Opdyke, N.D., Huang, K., Xu, G., Zhang, W.Y., Kent, D.V., 1987. Paleomagnetic results from the 2706 Silurian of the Yangtze paraplatform. Tectonophysics 139, 123–132. 2707 https://doi.org/10.1016/0040-1951(87)90201-0 2708 Palin, R.M., Santosh, M., Cao, W., Li, S.S., 2020. Secular change and the onset of plate tectonics on 2709 Earth. Earth-Sci. Rev. 2710 Pan, B., Brock, G.A., Skovsted, C.B., Betts, M.J., Topper, T.P., Li, G., 2018. Paterimitra pyramidalis 2711 Laurie, 1986, the first tommotiid discovered from the early Cambrian of North China. Gondwana 2712 Res. 63, 179–185. https://doi.org/10.1016/j.gr.2018.05.014 2713 Pandit, M.K., Carter, L.M., Ashwal, L.D., Tucker, R.D., Torsvik, T.H., Jamtveit, B., Bhushan, S.K., 2714 2003. Age, petrogenesis and significance of 1 Ga granitoids and related rocks from the Sendra 2715 area, Aravalli Craton, NW India. J. Asian Earth Sci. 22, 363–381. 2716 Paquette, J.L., Caby, R., Djouadi, M.T., Bouchez, J.L., 1998. U–Pb dating of the end of the Pan-African 2717 orogeny in the Tuareg shield: the post-collisional syn-shear Tioueine pluton (Western Hoggar, 2718 Algeria). Lithos 45, 245–253. https://doi.org/10.1016/S0024-4937(98)00034-6 2719 Parsons, A.J., Hosseini, K., Palin, R.M., Sigloch, K., 2020. Geological, geophysical and plate kinematic 2720 constraints for models of the India-Asia collision and the post-Triassic central Tethys oceans. 2721 Earth-Sci. Rev. 208, 103084. https://doi.org/10.1016/j.earscirev.2020.103084 2722 Pavlov, V., Bachtadse, V., Mikhailov, V., 2008. New Middle Cambrian and Middle Ordovician 2723 palaeomagnetic data from Siberia: Llandelian magnetostratigraphy and relative rotation between 2724 the Aldan and Anabar--Angara blocks. Earth Planet. Sci. Lett. 276, 229–242. 2725 Pavlov, V., Gallet, Y., 1998. Upper Cambrian to Middle Ordovician magnetostratigraphy from the 2726 Kulumbe river section (northwestern Siberia). Phys. Earth Planet. Inter. 108, 49–59. 2727 https://doi.org/10.1016/S0031-9201(98)00087-9 2728 Pavlov, V.E., Shatsillo, A.V., Petrov, P.Y., 2015. Paleomagnetism of the upper deposits in the 2729 Turukhansk and Olenek uplifts and Uda Pre-Sayan region and the neoproterozoic drift of the 2730 Siberian Platform. Izv. Phys. Solid Earth 51, 716–747. 2731 https://doi.org/10.1134/S1069351315050092 2732 Pavlov, V.E., Veselovskiy, R.V., Shatsillo, A.V., Gallet, Y., 2012. Magnetostratigraphy of the Ordovician 2733 Angara/Rozhkova River section: Further evidence for the Moyero reversed superchron. Izv. Phys. 2734 Solid Earth 48, 297–305. https://doi.org/10.1134/S1069351312040052 2735 Pedrosa-Soares, A.C., Noce, C.M., Wiedemann, C.M., Pinto, C.P., 2001. The Aracuaı-West-Congo 2736 Orogen in Brazil: an overview of a confined orogen formed during Gondwanaland assembly. 2737 Precambrian Res. 110, 307–323. 2738 Pehrsson, S.J., Berman, R.G., Eglington, B., Rainbird, R., 2013. Two supercontinents 2739 revisited: The case for a Rae family of cratons. Precambrian Res. 232, 27–43. 2740 https://doi.org/10.1016/j.precamres.2013.02.005 2741 Peng, Y., Yu, S., Li, S., Zhang, J., Liu, Y., Li, Y., Santosh, M., 2019. Early Neoproterozoic magmatic 2742 imprints in the Altun-Qilian-Kunlun region of the Qinghai-Tibet Plateau: Response to the 2743 assembly and breakup of Rodinia supercontinent. Earth-Sci. Rev. 199, 102954. 2744 https://doi.org/10.1016/j.earscirev.2019.102954 2745 Pisarevsky, S., Bylund, G., 1998. Neoproterozoic palaeomagnetic directions in rocks from a key section 2746 of the , southern Sweden. Geophys. J. Int. 133, 185–200. 2747 https://doi.org/10.1046/j.1365-246X.1998.1331497.x 2748 Pisarevsky, S.A., Elming, S.-Å., Pesonen, L.J., Li, Z.-X., 2014. Mesoproterozoic paleogeography: 2749 Supercontinent and beyond. Precambrian Res. 244, 207–225. 2750 https://doi.org/10.1016/j.precamres.2013.05.014 2751 Pisarevsky, S.A., Gladkochub, D.P., Konstantinov, K.M., Mazukabzov, A.M., Stanevich, A.M., Murphy, 2752 J.B., Tait, J.A., Donskaya, T.V., Konstantinov, I.K., 2013. Paleomagnetism of Cryogenian Kitoi 2753 mafic dykes in South Siberia: Implications for Neoproterozoic paleogeography. Precambrian Res. 2754 231, 372–382. https://doi.org/10.1016/j.precamres.2013.04.007 2755 Pisarevsky, S.A., Gurevich, E.L., Khramov, A.N., 1997. Palaeomagnetism of Lower Cambrian sediments 2756 from the Olenek River section (northern Siberia): palaeopoles and the problem of magnetic 2757 polarity in the Early Cambrian. Geophys. J. Int. 130, 746–756. https://doi.org/10.1111/j.1365- 2758 246X.1997.tb01869.x 2759 Pisarevsky, S.A., Murphy, J.B., Cawood, P.A., Collins, A.S., 2008. Late Neoproterozoic and Early 2760 Cambrian palaeogeography: models and problems. Geological Society, London, Special 2761 Publications 294, 9–31. https://doi.org/10.1144/SP294.2 2762 Pisarevsky, S.A., Natapov, L.M., 2003. Siberia and Rodinia. Tectonophysics 375, 221–245. 2763 https://doi.org/10.1016/j.tecto.2003.06.001 2764 Pisarevsky, S.A., Wingate, M.T.D., Stevens, M.K., Haines, P.W., 2007. Palaeomagnetic results from the 2765 Lancer 1 stratigraphic drillhole, Officer Basin, Western Australia, and implications for Rodinia 2766 reconstructions. Aust. J. Earth Sci. 54, 561–572. https://doi.org/10.1080/08120090701188962 2767 Plavsa, D., Collins, A.S., Foden, J.D., Clark, C., 2015. The evolution of a Gondwanan collisional orogen: 2768 A structural and geochronological appraisal from the Southern Granulite Terrane, South India. 2769 Tectonics 34, 820–857. 2770 Popov, V., Iosifidi, A., Khramov, A., Tait, J., Bachtadse, V., 2002. Paleomagnetism of Upper Vendian 2771 sediments from the Winter Coast, White Sea region, Russia: implications for the paleogeography 2772 of Baltica during Neoproterozoic times. J. Geophys. Res. [Solid Earth] 107, EPM-10. 2773 Popov, V.V., Khramov, A.N., Bachtadse, V., 2005. Palaeomagnetism, magnetic stratigraphy, and 2774 petromagnetism of the Upper Vendian sedimentary rocks in the sections of the Zolotitsa River 2775 and in the Verkhotina Hole, Winter Coast of the White Sea, Russia. Russian Journal of Earth 2776 Sciences 7. 2777 Powell, C.M., Pisarevsky, S.A., 2002. Late Neoproterozoic assembly of East Gondwana. Geology 30, 3– 2778 6. 2.0.CO;2">https://doi.org/10.1130/0091-7613(2002)030<0003:LNAOEG>2.0.CO;2 2779 Powerman, V., Shatsillo, A., Coe, R., Zhao, X., Gladkochub, D., Buchwaldt, R., Pavlov, V., 2013. 2780 Palaeogeography of the Siberian platform during middle Palaeozoic Times ( 450--400 Ma): new 2781 palaeomagnetic evidence from the Lena and Nyuya rivers. Geophys. J. Int. 194, 1412–1440. 2782 Qiantao, B., Shanlin, G., Dihui, L., Zhengren, Y., Chengfa, C., Xiaoquan, L., 2001. A study of the 2783 Kunlun-Qilian-Qinling suture system. Acta Geologica Sinica-English Edition 75, 364–374. 2784 Rapalini, A.E., 2006. New late Proterozoic paleomagnetic pole for the Rio de la Plata craton: Implications 2785 for Gondwana. Precambrian Res. 147, 223–233. https://doi.org/10.1016/j.precamres.2006.01.016 2786 Rapalini, A.E., Cingolani, C.A., 2004. First Late Ordovician Paleomagnetic Pole for the Cuyania 2787 (Precordillera) Terrane of Western Argentina: a Microcontinent or a Laurentian Plateau. 2788 Gondwana Res. 7, 1089–1104. https://doi.org/10.1016/S1342-937X(05)71086-8 2789 Rapalini, A.E., Tohver, E., Bettucci, L.S., Lossada, A.C., Barcelona, H., Pérez, C., 2015. The late 2790 Neoproterozoic Sierra de las Ánimas Magmatic Complex and Playa , 2791 southern Uruguay, revisited: Paleogeographic implications of new paleomagnetic and precise 2792 geochronologic data. Precambrian Res. 259, 143–155. 2793 https://doi.org/10.1016/j.precamres.2014.11.021 2794 Raub, T.D., Kirschvink, J.L., Evans, D.A.D., 2007. True polar wander: Linking deep and shallow 2795 geodynamics to hydro-and biospheric hypotheses. Treatise on geophysics 5, 565–589. 2796 Raza, M., Khan, A., Bhardwaj, V.R., Rais, S., 2012. Geochemistry of Mesoproterozoic sedimentary rocks 2797 of upper Vindhyan Group, southeastern Rajasthan and implications for weathering history, 2798 composition and tectonic setting of continental crust in the northern part of Indian shield. J. Asian 2799 Earth Sci. 48, 160–172. 2800 Robert, B., Besse, J., Blein, O., Greff-Lefftz, M., Baudin, T., Lopes, F., Meslouh, S., Belbadaoui, M., 2801 2017. Constraints on the Ediacaran inertial interchange true polar wander hypothesis: A new 2802 paleomagnetic study in Morocco (West African Craton). Precambrian Res. 295, 90–116. 2803 https://doi.org/10.1016/j.precamres.2017.04.010 2804 Robert, B., Domeier, M., Jakob, J., 2020. Iapetan Oceans: An analog of Tethys? Geology. 2805 Rodionov, V.P., 1966. Dipole character of the Geomagnetic field in the Late Cambrian and the 2806 Ordovician in the south of the Siberian Platform. Geol. Geofiz. 1, 94–101. 2807 Rodionov, V.P., Dekkers, M.J., Khramov, A.N., Gurevich, E.L., Krijgsman, W., Duermeijer, C.E., 2808 Heslop, D., 2003. Paleomagnetism and cyclostratigraphy of the Middle Ordovician Krivolutsky 2809 suite, Krivaya Luka section, southern Siberian platform: record of non-synchronous NRM- 2810 components or a non-axial geomagnetic field? Studia Geophys. Geodaetica 47, 255–274. 2811 Rodionov, V.P., Khramov, A.N., Pisarevskii, S.A., Popov, V.V., Iosifidi, A.G., 1998. Geomagnetic 2812 Reversals in the Early Paleozoic. 1. A Late Cambrian Reversal Recorded in the Ichera Section, 2813 Southern Siberian Platform. Izv. Phys. Solid Earth 34, 1009–1017. 2814 Rogers, J.J.W., Santosh, M., 2002. Configuration of Columbia, a Mesoproterozoic Supercontinent. 2815 Gondwana Res. 5, 5–22. https://doi.org/10.1016/S1342-937X(05)70883-2 2816 Ross, M.I., Scotese, C.R., 1988. A hierarchical tectonic model of the and Caribbean 2817 region. Tectonophysics. 2818 Ruppel, A., Jacobs, J., Eagles, G., Läufer, A., Jokat, W., 2018. New geophysical data from a key region in 2819 East Antarctica: Estimates for the spatial extent of the Tonian Oceanic Arc Super Terrane 2820 (TOAST). Gondwana Res. 59, 97–107. https://doi.org/10.1016/j.gr.2018.02.019 2821 Safonova, I., Biske, G., Romer, R.L., Seltmann, R., Simonov, V., Maruyama, S., 2016. Middle Paleozoic 2822 mafic magmatism and ocean plate stratigraphy of the South Tianshan, Kyrgyzstan. Gondwana 2823 Res. 30, 236–256. https://doi.org/10.1016/j.gr.2015.03.006 2824 Schellart, W.P., Lister, G.S., Toy, V.G., 2006. A Late Cretaceous and Cenozoic reconstruction of the 2825 Southwest Pacific region: Tectonics controlled by subduction and slab rollback processes. Earth- 2826 Sci. Rev. 76, 191–233. https://doi.org/10.1016/j.earscirev.2006.01.002 2827 Scherstén, A., Årebäck, H., Cornell, D., Hoskin, P., Åberg, A., Armstrong, R., 2000. Dating mafic– 2828 ultramafic intrusions by ion-microprobing contact-melt zircon: examples from SW Sweden. 2829 Contrib. Mineral. Petrol. 139, 115–125. https://doi.org/10.1007/s004100050577 2830 Schmidt, P.W., Williams, G.E., 2010. Ediacaran palaeomagnetism and apparent polar wander path for 2831 Australia: no large true polar wander. Geophys. J. Int. 182, 711–726. 2832 https://doi.org/10.1111/j.1365-246X.2010.04652.x 2833 Schmidt, P.W., Williams, G.E., 1996. Palaeomagnetism of the ejecta-bearing Bunyeroo Formation, late 2834 Neoproterozoic, Adelaide fold belt, and the age of the Acraman impact. Earth Planet. Sci. Lett. 2835 144, 347–357. https://doi.org/10.1016/S0012-821X(96)00169-0 2836 Schmidt, P.W., Williams, G.E., 1995. The Neoproterozoic climatic paradox: Equatorial palaeolatitude for 2837 near sea level in South Australia. Earth Planet. Sci. Lett. 134, 107–124. 2838 https://doi.org/10.1016/0012-821X(95)00106-M 2839 Schmidt, P.W., Williams, G.E., Embleton, B.J.J., 1991. Low palaeolatitude of Late Proterozoic 2840 glaciation: early timing of remanence in haematite of the Elatina Formation, South Australia. 2841 Earth Planet. Sci. Lett. 105, 355–367. https://doi.org/10.1016/0012-821X(91)90177-J 2842 Schmidt, P.W., Williams, G.E., McWilliams, M.O., 2009. Palaeomagnetism and magnetic anisotropy of 2843 late Neoproterozoic strata, South Australia: Implications for the palaeolatitude of late Cryogenian 2844 glaciation, cap carbonate and the Ediacaran System. Precambrian Res. 174, 35–52. 2845 https://doi.org/10.1016/j.precamres.2009.06.002 2846 Scotese, C.R., 2016. PALEOMAP PaleoAtlas for GPlates and the PaleoData Plotter Program, 2847 PALEOMAP Project. See http://www. earthbyte. org/paleomap-paleoatlas-for-gplates. 2848 Scotese, C.R., 2004. A Continental Drift Flipbook. J. Geol. 112, 729–741. https://doi.org/10.1086/424867 2849 Seton, M., Müller, R.D., Zahirovic, S., Gaina, C., Torsvik, T., Shephard, G., Talsma, A., Gurnis, M., 2850 Turner, M., Maus, S., Chandler, M., 2012. Global continental and ocean basin reconstructions 2851 since 200 Ma. Earth-Sci. Rev. 113, 212–270. https://doi.org/10.1016/j.earscirev.2012.03.002 2852 Shapiro, M.N., Solov’ev, A.V., 2009. Formation of the Olyutorsky–Kamchatka foldbelt: a kinematic 2853 model. Russ. Geol. Geophys. 50, 668–681. https://doi.org/10.1016/j.rgg.2008.10.006 2854 Shatsillo, A.V., Paverman, V.I., Pavlov, V.E., 2007. Middle paleozoic segment of the apparent polar 2855 wander path from the Siberian platform: New paleomagnetic evidence for the Silurian of the 2856 Nyuya-Berezovskii facial province. Izv. Phys. Solid Earth 43, 880–889. 2857 https://doi.org/10.1134/S1069351307100102 2858 Shephard, G.E., Müller, R.D., Seton, M., 2013. The tectonic evolution of the Arctic since Pangea 2859 breakup: Integrating constraints from surface geology and geophysics with mantle structure. 2860 Earth-Sci. Rev. 124, 148–183. https://doi.org/10.1016/j.earscirev.2013.05.012 2861 Shervais, J.W., Dennis, A.J., McGee, J.J., Secor, D., 2003. Deep in the heart of Dixie: Pre-Alleghanian 2862 eclogite and HP granulite metamorphism in the Carolina Terrane, South Carolina, USA. J. 2863 Metamorph. Geol. 21, 65–80. 2864 Shi, R., Yang, J., Wu, C., Wooden, J., 2004. First SHRIMP dating for the formation of the late Sinian 2865 Yushigou Ophiolite, North Qilian Mountains. ACTA GEOLOGICA SINICA-CHINESE 2866 EDITION- 78, 649–657. 2867 Shi, Y., Zhang, W., Kröner, A., Li, L., Jian, P., 2018. Cambrian ophiolite complexes in the Beishan area, 2868 China, southern margin of the Central Asian Orogenic Belt. J. Asian Earth Sci. 153, 193–205. 2869 https://doi.org/10.1016/j.jseaes.2017.05.021 2870 Sigloch, K., Mihalynuk, M.G., 2013. Intra-oceanic subduction shaped the assembly of Cordilleran North 2871 America. Nature 496, 50–56. https://doi.org/10.1038/nature12019 2872 Slagstad, T., Kulakov, E., Kirkland, C.L., Roberts, N.M.W., Ganerød, M., 2019. Breaking the Grenville– 2873 Sveconorwegian link in Rodinia reconstructions. Terra Nova 31, 430–437. 2874 https://doi.org/10.1111/ter.12406 2875 Smethurst, M.A., Khramov, A.N., 1992. A new Devonian palaeomagnetic pole for the Russian platform 2876 and Baltica, and related apparent polar wander. Geophys. J. Int. 108, 179–192. 2877 https://doi.org/10.1111/j.1365-246X.1992.tb00848.x 2878 Smethurst, M.A., Khramov, A.N., Pisarevsky, S., 1998. Palaeomagnetism of the Lower Ordovician 2879 Orthoceras Limestone, St. Petersburg, and a revised drift history for Baltica in the early 2880 Palaeozoic. Geophys. J. Int. 133, 44–56. https://doi.org/10.1046/j.1365-246X.1998.1331463.x 2881 Smirnov, A.V., Evans, D.A.D., Ernst, R.E., Söderlund, U., Li, Z.-X., 2013. Trading partners: Tectonic 2882 ancestry of southern Africa and western Australia, in Archean supercratons and 2883 Zimgarn. Precambrian Res. 224, 11–22. https://doi.org/10.1016/j.precamres.2012.09.020 2884 Sohl, L.E., Christie-Blick, N., Kent, D.V., 1999. Paleomagnetic polarity reversals in Marinoan (ca. 600 2885 Ma) glacial deposits of Australia: Implications for the duration of low-latitude glaciation in 2886 Neoproterozoic time. GSA Bulletin 111, 1120–1139. 2.3.CO;2">https://doi.org/10.1130/0016- 2887 7606(1999)111<1120:PPRIMC>2.3.CO;2 2888 Soldati, G., Boschi, L., Piersanti, A., Spada, G., 2001. The effect of global seismicity on the polar motion 2889 of a viscoelastic Earth. J. Geophys. Res. 106, 6761–6767. https://doi.org/10.1029/2000JB900354 2890 Song, S., Li, X.-H., 2019. A positive test for the Greater Tarim Block at the heart of Rodinia: Mega- 2891 dextral suturing of supercontinent assembly: COMMENT. Geology 47, e453–e453. 2892 Song, S., Niu, Y., Su, L., Xia, X., 2013. Tectonics of the North Qilian orogen, NW China. Gondwana 2893 Res. 23, 1378–1401. https://doi.org/10.1016/j.gr.2012.02.004 2894 Song, S., Niu, Y., Zhang, L., Wei, C., Liou, J.G., Su, L., 2009. Tectonic evolution of early Paleozoic HP 2895 metamorphic rocks in the North Qilian Mountains, NW China: New perspectives. J. Asian Earth 2896 Sci. 35, 334–353. https://doi.org/10.1016/j.jseaes.2008.11.005 2897 Song, S., Su, L., Li, X.-H., Niu, Y., Zhang, L., 2012. Grenville-age orogenesis in the Qaidam-Qilian 2898 block: The link between South China and Tarim. Precambrian Res. 220–221, 9–22. 2899 https://doi.org/10.1016/j.precamres.2012.07.007 2900 Stampfli, G.M., Borel, G.D., 2002. A plate tectonic model for the Paleozoic and Mesozoic constrained by 2901 dynamic plate boundaries and restored synthetic oceanic isochrons. Earth Planet. Sci. Lett. 196, 2902 17–33. https://doi.org/10.1016/S0012-821X(01)00588-X 2903 Stearn, J.E.F., Piper, J.D.A., 1984. Sub-Department of Geophysics, University of Liverpool, Liverpool 2904 L69 3BX (Gt. Britain). Precambrian Res. 23, 201–246. 2905 Steinberger, B., Torsvik, T.H., 2008. Absolute plate motions and true polar wander in the absence of 2906 hotspot tracks. Nature 452, 620–623. https://doi.org/10.1038/nature06824 2907 Stern, R.J., 1994. Arc Assembly and in the Neoproterozoic East African Orogen: 2908 Implications for the Consolidation of Gondwanaland. Annu. Rev. Earth Planet. Sci. 22, 319–351. 2909 https://doi.org/10.1146/annurev.ea.22.050194.001535 2910 Stern, R.J., Miller, N.R., 2018. Did the transition to plate tectonics cause Neoproterozoic Snowball Earth? 2911 Terra Nova 30, 87–94. https://doi.org/10.1111/ter.12321 2912 Strachan, R.A., Collins, A.S., Buchan, C., Nance, R.D., Murphy, J.B., D’Lemos, R.S., 2007. Terrane 2913 analysis along a Neoproterozoic active margin of Gondwana: insights from U–Pb zircon 2914 geochronology. J. Geol. Soc. London 164, 57–60. https://doi.org/10.1144/0016-76492006-014 2915 Sun, L.-S., Huang, B.-C., 2009. New paleomagnetic result for Ordovician rocks from the Tarim Block, 2916 Northwest China and its tectonic implications. 2917 Surkis, Y.F., Rodionov, V.P., Khramov, A.N., Gurevich, E.L., Westfahl, M., 1999. Geomagnetic 2918 reversals in the early Paleozoic: 3. Reversals recorded in redbeds of the lower Ordovician Mandra 2919 section, Siberia. Izv. Phys. Solid Earth 35, 347–357. 2920 Sutherland, R., Dickens, G.R., Blum, P., Agnini, C., Alegret, L., Asatryan, G., Bhattacharya, J., 2921 Bordenave, A., Chang, L., Collot, J., Cramwinckel, M.J., Dallanave, E., Drake, M.K., Etienne, 2922 S.J.G., Giorgioni, M., Gurnis, M., Harper, D.T., Huang, H.-H.M., Keller, A.L., Lam, A.R., Li, H., 2923 Matsui, H., Morgans, H.E.G., Newsam, C., Park, Y.-H., Pascher, K.M., Pekar, S.F., Penman, 2924 D.E., Saito, S., Stratford, W.R., Westerhold, T., Zhou, X., 2020. Continental-scale geographic 2925 change across during subduction initiation. Geology 48, 419–424. 2926 https://doi.org/10.1130/G47008.1 2927 Swanson-Hysell, N.L., Maloof, A.C., Kirschvink, J.L., Evans, D.A.D., Halverson, G.P., Hurtgen, M.T., 2928 2012. Constraints on Neoproterozoic paleogeography and Paleozoic orogenesis from 2929 paleomagnetic records of the Bitter Springs Formation, Amadeus Basin, central Australia. Am. J. 2930 Sci. 312, 817–884. https://doi.org/10.2475/08.2012.01 2931 Tetley, M.G., 2018. Constraining Earth’s plate tectonic evolution through data mining and knowledge 2932 discovery (PhD). The University of Sydney. 2933 Tetley, M.G., Williams, S.E., Gurnis, M., Flament, N., Müller, R.D., 2019. Constraining absolute plate 2934 motions since the Triassic. J. Geophys. Res. [Solid Earth]. https://doi.org/10.1029/2019JB017442 2935 Thomas, R.J., De Waele, B., Schofield, D.I., Goodenough, K.M., Horstwood, M., Tucker, R., Bauer, W., 2936 Annells, R., Howard, K., Walsh, G., Rabarimanana, M., Rafahatelo, J.M., Ralison, A.V., 2937 Randriamananjara, T., 2009. Geological evolution of the Neoproterozoic Bemarivo Belt, northern 2938 Madagascar. Precambrian Res. 172, 279–300. https://doi.org/10.1016/j.precamres.2009.04.008 2939 Thompson, M.D., Grunow, A.M., Ramezani, J., 2010. Cambro-Ordovician paleogeography of the 2940 Southeastern New England Avalon Zone: Implications for Gondwana breakup. GSA Bulletin 2941 122, 76–88. https://doi.org/10.1130/B26581.1 2942 Thorogood, E.J., 1990. Provenance of the pre-Devonian sediments of England and Wales: Sm-Nd 2943 isotopic evidence. J. Geol. Soc. London 147, 591–594. https://doi.org/10.1144/gsjgs.147.4.0591 2944 Torsvik, T.H., Ashwal, L.D., Tucker, R.D., Eide, E.A., 2001. Neoproterozoic geochronology and 2945 palaeogeography of the Seychelles microcontinent: the India link. Precambrian Res. 110, 47–59. 2946 https://doi.org/10.1016/S0301-9268(01)00180-2 2947 Torsvik, T.H., Burke, K., Steinberger, B., Webb, S.J., Ashwal, L.D., 2010a. sampled by 2948 plumes from the core-mantle boundary. Nature 466, 352–355. 2949 https://doi.org/10.1038/nature09216 2950 Torsvik, T.H., Cocks, L.R.M., 2017. The integration of palaeomagnetism, the geological record and 2951 mantle tomography in the location of ancient continents. Geol. Mag. 1–19. 2952 https://doi.org/10.1017/S001675681700098X 2953 Torsvik, T.H., Cocks, L.R.M., 2016. Earth History and Palaeogeography. Cambridge University Press. 2954 Torsvik, T.H., Cocks, L.R.M., 2009. The Lower Palaeozoic palaeogeographical evolution of the 2955 northeastern and eastern peri-Gondwanan margin from Turkey to New Zealand. Geological 2956 Society, London, Special Publications 325, 3–21. https://doi.org/10.1144/SP325.2 2957 Torsvik, T.H., Dietmar Muller, R., Van der Voo, R., Steinberger, B., Gaina, C., 2008. Global Plate 2958 Motion Frames: Toward a Unified Model. Reviews of Geophysics 46. 2959 https://doi.org/10.1029/2007RG000227 2960 Torsvik, T.H., Domeier, M., 2017. Correspondence: Numerical modelling of the PERM anomaly and the 2961 Emeishan large igneous province. Nat. Commun. 8, 821. https://doi.org/10.1038/s41467-017- 2962 00125-2 2963 Torsvik, T.H., Rehnström, E.F., 2001. Cambrian palaeomagnetic data from Baltica: implications for true 2964 polar wander and Cambrian palaeogeography. J. Geol. Soc. London 158, 321–329. 2965 https://doi.org/10.1144/jgs.158.2.321 2966 Torsvik, T.H., Smethurst, M.A., 1989. GMAP32—Geographic Mapping and Reconstruction System. 2967 Geol. Surv. of Norway, Oslo 1997. 2968 Torsvik, T.H., Smethurst, M.A., Meert, J.G., Van der Voo, R., McKerrow, W.S., Brasier, Sturt, B.A., 2969 Walderhaug, H.J., 1996. Continental break-up and collision in the Neoproterozoic and Palaeozoic 2970 — A tale of Baltica and Laurentia. Earth-Sci. Rev. 40, 229–258. https://doi.org/10.1016/0012- 2971 8252(96)00008-6 2972 Torsvik, T.H., Steinberger, B., Gurnis, M., Gaina, C., 2010b. Plate tectonics and net lithosphere rotation 2973 over the past 150My. Earth Planet. Sci. Lett. 291, 106–112. 2974 https://doi.org/10.1016/j.epsl.2009.12.055 2975 Torsvik, T.H., Tait, J., Moralev, V.M., McKerrow, W.S., Sturt, B.A., Roberts, D., 1995. Ordovician 2976 palaeogeography of Siberia and adjacent continents. J. Geol. Soc. London 152, 279–287. 2977 Torsvik, T.H., Trench, A., 1991. The Lower—Middle Ordovician palaeofield of : southern 2978 Sweden “revisited.” Phys. Earth Planet. Inter. 65, 283–291. 2979 Torsvik, T.H., Trench, A., McKerrow, W.S., 1994. Implications of palaeomagnetic data from the 2980 Tortworth Silurian inlier (southern Britain) to palaeogeography and Variscan tectonism. Geophys. 2981 J. Int. 119, 91–100. 2982 Torsvik, T.H., Trench, A., Svensson, I., Walderhaug, H.J., 1993. Palaeogeographic significance of mid- 2983 Silurian palaeomagnetic results from southern Britain—major revision of the apparent polar 2984 wander path for eastern Avalonia. Geophys. J. Int. 113, 651–668. 2985 Torsvik, T.H., Van der Voo, R., 2002. Refining Gondwana and Pangea palaeogeography: estimates of 2986 Phanerozoic non-dipole (octupole) fields. Geophys. J. Int. 151, 771–794. 2987 https://doi.org/10.1046/j.1365-246X.2002.01799.x 2988 Torsvik, T.H., van der Voo, R., Doubrovine, P.V., Burke, K., Steinberger, B., Ashwal, L.D., Trønnes, 2989 R.G., Webb, S.J., Bull, A.L., 2014. Deep mantle structure as a reference frame for movements in 2990 and on the Earth. Proc. Natl. Acad. Sci. U. S. A. 111, 8735–8740. 2991 https://doi.org/10.1073/pnas.1318135111 2992 Torsvik, T.H., Van der Voo, R., Preeden, U., Mac Niocaill, C., Steinberger, B., Doubrovine, P.V., van 2993 Hinsbergen, D.J.J., Domeier, M., Gaina, C., Tohver, E., Meert, J.G., McCausland, P.J.A., Cocks, 2994 L.R.M., 2012. Phanerozoic polar wander, palaeogeography and dynamics. Earth-Sci. Rev. 114, 2995 325–368. https://doi.org/10.1016/j.earscirev.2012.06.007 2996 Trench, A., Torsvik, T.H., 1991. A revised Palaeozoic apparent polar wander path for Southern Britain 2997 (Eastern Avalonia). Geophys. J. Int. 104, 227–233. https://doi.org/10.1111/j.1365- 2998 246X.1991.tb02506.x 2999 Trench, A., Torsvik, T.H., Dentith, M.C., Walderhaug, H., Traynor, J.-J., 1992. A high southerly 3000 palaeolatitude for Southern Britain in Early Ordovician times: palaeomagnetic data from the 3001 Treffgarne Volcanic Formation SW Wales. Geophys. J. Int. 108, 89–100. 3002 Tretyakov, A.A., Degtyarev, K.E., Kovach, V.P., Kotov, A.B., Salnikova, E.B., Pilitsyna, A.V., 3003 Yakovleva, S.Z., 2016. The migmatite–gneiss complex of the Chuya–Kendyktas sialic massif 3004 (Southern Kazakhstan): Structure and age. Dokl. Earth Sci. 467, 236–240. 3005 https://doi.org/10.1134/S1028334X16030156 3006 Tucker, R.D., Ashwal, L.D., Torsvik, T.H., 2001. U–Pb geochronology of Seychelles granitoids: a 3007 Neoproterozoic continental arc fragment. Earth Planet. Sci. Lett. 187, 27–38. 3008 https://doi.org/10.1016/S0012-821X(01)00282-5 3009 Tucker, R.D., Roig, J.Y., Macey, P.H., Delor, C., Amelin, Y., Armstrong, R.A., Rabarimanana, M.H., 3010 Ralison, A.V., 2011. A new geological framework for south-central Madagascar, and its 3011 relevance to the “out-of-Africa” hypothesis. Precambrian Res. 185, 109–130. 3012 https://doi.org/10.1016/j.precamres.2010.12.008 3013 Tung, K., Yang, H.-J., Yang, H.-Y., Liu, D., Zhang, J., Wan, Y., Tseng, C.-Y., 2007. SHRIMP U-Pb 3014 geochronology of the zircons from the Precambrian basement of the Qilian Block and its 3015 geological significances. Chin. Sci. Bull. 52, 2687–2701. https://doi.org/10.1007/s11434-007- 3016 0356-0 3017 Vaes, B., Van Hinsbergen, D.J.J., Boschman, L.M., 2019. Reconstruction of subduction and back-arc 3018 spreading in the NW Pacific and Aleutian Basin: Clues to causes of Cretaceous and Eocene plate 3019 reorganizations. Tectonics 38, 1367–1413. 3020 van der Meer, D.G., Spakman, W., van Hinsbergen, D.J.J., Amaru, M.L., Torsvik, T.H., 2010. Towards 3021 absolute plate motions constrained by lower-mantle slab remnants. Nat. Geosci. 3, 36–40. 3022 https://doi.org/10.1038/ngeo708 3023 Van der Voo, R., 1990. The reliability of paleomagnetic data. Tectonophysics 184, 1–9. 3024 https://doi.org/10.1016/0040-1951(90)90116-P 3025 Van der Voo, R., French, R.B., Williams, D.W., 1976. Paleomagnetism of the Wilberns Formation 3026 (Texas) and the late Cambrian paleomagnetic field for North America. J. Geophys. Res. 81, 3027 5633–5638. 3028 van Hinsbergen, D.J.J., Lippert, P.C., Dupont-Nivet, G., McQuarrie, N., Doubrovine, P.V., Spakman, W., 3029 Torsvik, T.H., 2012. Greater India Basin hypothesis and a two-stage Cenozoic collision between 3030 India and Asia. Proc. Natl. Acad. Sci. U. S. A. 109, 7659–7664. 3031 https://doi.org/10.1073/pnas.1117262109 3032 van Hinsbergen, D.J.J., Torsvik, T.H., Schmid, S.M., Maţenco, L.C., Maffione, M., Vissers, R.L.M., 3033 Gürer, D., Spakman, W., 2020. Orogenic architecture of the Mediterranean region and kinematic 3034 reconstruction of its tectonic evolution since the Triassic. Gondwana Res. 81, 79–229. 3035 https://doi.org/10.1016/j.gr.2019.07.009 3036 Van Lente, B., Ashwal, L.D., Pandit, M.K., Bowring, S.A., Torsvik, T.H., 2009. Neoproterozoic 3037 hydrothermally altered basaltic rocks from Rajasthan, northwest India: Implications for late 3038 Precambrian tectonic evolution of the Aravalli Craton. Precambrian Res. 170, 202–222. 3039 https://doi.org/10.1016/j.precamres.2009.01.007 3040 van Staal, C.R., Barr, S.M., Brendan Murphy, J., 2012. Provenance and tectonic evolution of Ganderia: 3041 Constraints on the evolution of the Iapetus and Rheic oceans. Geology 40, 987–990. 3042 https://doi.org/10.1130/G33302.1 3043 Vernikovsky, V.A., Vernikovskaya, A.E., 2001. Central Taimyr accretionary belt (Arctic Asia): Meso– 3044 Neoproterozoic tectonic evolution and Rodinia breakup. Precambrian Res. 110, 127–141. 3045 https://doi.org/10.1016/S0301-9268(01)00184-X 3046 Vernikovsky, V.A., Vernikovskaya, A.E., Kotov, A.B., Sal’nikova, E.B., Kovach, V.P., 2003. 3047 Neoproterozoic accretionary and collisional events on the western margin of the Siberian craton: 3048 new geological and geochronological evidence from the Yenisey Ridge. Tectonophysics 375, 3049 147–168. https://doi.org/10.1016/S0040-1951(03)00337-8 3050 Vernikovsky, V.A., Vernikovskaya, A.E., Pease, V.L., Gee, D.G., 2004. Neoproterozoic Orogeny along 3051 the margins of Siberia. Geological Society, London, Memoirs 30, 233–248. 3052 https://doi.org/10.1144/GSL.MEM.2004.030.01.18 3053 Veselovskiy, R.V., Arzamastsev, A.A., 2011. Evidence for the Mesozoic endogenous activity in the 3054 northeastern part of the Fennoscandian Shield. Dokl. Earth Sci. 438, 754–758. 3055 https://doi.org/10.1134/S1028334X11060377 3056 Vick, H.K., Channell, J.E.T., Opdyke, N.D., 1987. Ordovician docking of the Carolina Slate Belt: 3057 Paleomagnetic data. Tectonics 6, 573–583. https://doi.org/10.1029/TC006i005p00573 3058 Vizan, H., Carney, J.N., Turner, P., Ixer, R.A., Tomasso, M., Mullen, R.P., Clarke, P., 2003. Late 3059 Neoproterozoic to Early Palaeozoic palaeogeography of Avalonia: some palaeomagnetic 3060 constraints from Nuneaton, central England. Geol. Mag. 140, 685–705. 3061 https://doi.org/10.1017/S001675680300832X 3062 von Raumer, J.F., Stampfli, G.M., 2008. The birth of the Rheic Ocean—Early Palaeozoic subsidence 3063 patterns and subsequent tectonic plate scenarios. Tectonophysics 461, 9–20. 3064 Walderhaug, H.J., Torsvik, T.H., Eide, E.A., Sundvoll, B., Bingen, B., 1999. Geochronology and 3065 palaeomagnetism of the Hunnedalen dykes, SW Norway: implications for the Sveconorwegian 3066 apparent polar wander loop. Earth Planet. Sci. Lett. 169, 71–83. https://doi.org/10.1016/S0012- 3067 821X(99)00066-7 3068 Walderhaug, H.J., Torsvik, T.H., Halvorsen, E., 2007. The Egersund dykes (SW Norway): a robust Early 3069 Ediacaran (Vendian) palaeomagnetic pole from Baltica. Geophys. J. Int. 168, 935–948. 3070 https://doi.org/10.1111/j.1365-246X.2006.03265.x 3071 Wang, C., Liu, L., Wang, Y.-H., He, S.-P., Li, R.-S., Li, M., Yang, W.-Q., Cao, Y.-T., Collins, A.S., Shi, 3072 C., Wu, Z.-N., 2015. Recognition and tectonic implications of an extensive Neoproterozoic 3073 volcano-sedimentary rift basin along the southwestern margin of the Tarim Craton, northwestern 3074 China. Precambrian Res. 257, 65–82. https://doi.org/10.1016/j.precamres.2014.11.022 3075 Wang, C.-C., Jacobs, J., Elburg, M.A., Läufer, A., Thomas, R.J., Elvevold, S., 2020. Grenville-age 3076 continental arc magmatism and crustal evolution in central Dronning Maud Land (East 3077 Antarctica): Zircon geochronological and HfO isotopic evidence. Gondwana Res. 3078 https://doi.org/10.1016/j.gr.2019.12.004 3079 Weil, A.B., Geissman, J.W., Ashby, J.M., 2006. A new paleomagnetic pole for the Neoproterozoic Uinta 3080 Mountain supergroup, Central Rocky Mountain States, USA. Precambrian Res. 147, 234–259. 3081 https://doi.org/10.1016/j.precamres.2006.01.017 3082 Weil, A.B., Geissman, J.W., Van der Voo, R., 2004. Paleomagnetism of the Neoproterozoic Chuar 3083 Group, Grand Canyon Supergroup, Arizona: implications for Laurentia’s Neoproterozoic APWP 3084 and Rodinia break-up. Precambrian Res. 129, 71–92. 3085 Weil, A.B., Van der Voo, R., Mac Niocaill, C., Meert, J.G., 1998. The Proterozoic supercontinent 3086 Rodinia: paleomagnetically derived reconstructions for 1100 to 800 Ma. Earth Planet. Sci. Lett. 3087 154, 13–24. https://doi.org/10.1016/S0012-821X(97)00127-1 3088 Wen, B., Evans, D.A.D., Li, Y.-X., 2017. Neoproterozoic paleogeography of the Tarim Block: An 3089 extended or alternative “missing-link” model for Rodinia? Earth Planet. Sci. Lett. 458, 92–106. 3090 https://doi.org/10.1016/j.epsl.2016.10.030 3091 Wen, B., Evans, D.A.D., Wang, C., Li, Y.-X., Jing, X., 2019. A positive test for the Greater Tarim Block 3092 at the heart of Rodinia: Mega-dextral suturing of supercontinent assembly: REPLY. Geology 47, 3093 e454–e454. 3094 Wen, B., Evans, D.A.D., Wang, C., Li, Y.-X., Jing, X., 2018. A positive test for the Greater Tarim Block 3095 at the heart of Rodinia: Mega-dextral suturing of supercontinent assembly. Geology 46, 687–690. 3096 https://doi.org/10.1130/G40254.1 3097 Wen, B., Li, Y.-X., Zhu, W., 2013. Paleomagnetism of the Neoproterozoic diamictites of the Qiaoenbrak 3098 formation in the Aksu area, NW China: Constraints on the paleogeographic position of the Tarim 3099 Block. Precambrian Res. 226, 75–90. https://doi.org/10.1016/j.precamres.2012.10.018 3100 White, C.E., Barr, S.M., Hamilton, M.A., Murphy, J.B., 2020. Age and tectonic setting of Neoproterozoic 3101 granitoid rocks, Antigonish Highlands, Nova Scotia, Canada: Implications for Avalonia in the 3102 northern Appalachian orogen. Can. J. Earth Sci. 3103 White, C.E., Barr, S.M., Miller, B.V., Hamilton, M.A., 2002. Granitoid plutons of the Brookville terrane, 3104 southern : petrology, age, and tectonic setting. 3105 Wilhem, C., Windley, B.F., Stampfli, G.M., 2012. The Altaids of Central Asia: A tectonic and 3106 evolutionary innovative review. Earth-Sci. Rev. 113, 303–341. 3107 https://doi.org/10.1016/j.earscirev.2012.04.001 3108 Williams, G.E., Schmidt, P.W., 2015. Low paleolatitude for the late Cryogenian interglacial succession, 3109 South Australia: paleomagnetism of the Angepena Formation, Adelaide Geosyncline. Aust. J. 3110 Earth Sci. 62, 243–253. https://doi.org/10.1080/08120099.2015.1003967 3111 Williams, S., Flament, N., Dietmar Müller, R., Butterworth, N., 2015. Absolute plate motions since 130 3112 Ma constrained by subduction zone kinematics. Earth Planet. Sci. Lett. 418, 66–77. 3113 https://doi.org/10.1016/j.epsl.2015.02.026 3114 Windley, B.F., Alexeiev, D., Xiao, W., Kröner, A., Badarch, G., 2007. Tectonic models for accretion of 3115 the Central Asian Orogenic Belt. J. Geol. Soc. London 164, 31–47. https://doi.org/10.1144/0016- 3116 76492006-022 3117 Wingate, M.T.D., Giddings, J.W., 2000. Age and palaeomagnetism of the Mundine Well dyke swarm, 3118 Western Australia: implications for an Australia–Laurentia connection at 755 Ma. Precambrian 3119 Res. 100, 335–357. https://doi.org/10.1016/S0301-9268(99)00080-7 3120 Wingate, M.T.D., Pisarevsky, S.A., De Waele, B., 2010. Paleomagnetism of the 765 Ma Luakela 3121 volcanics in Northwest Zambia and implications for Neoproterozoic positions of the Congo 3122 Craton. Am. J. Sci. 310, 1333–1344. https://doi.org/10.2475/10.2010.05 3123 Wu, C., Zuza, A.V., Yin, A., Liu, C., Reith, R.C., Zhang, J., Liu, W., Zhou, Z., 2017. Geochronology and 3124 geochemistry of Neoproterozoic granitoids in the central Qilian Shan of northern Tibet: 3125 Reconstructing the amalgamation processes and tectonic history of Asia. Lithosphere 9, 609–636. 3126 https://doi.org/10.1130/L640.1 3127 Wu, J., Suppe, J., Lu, R., Kanda, R., 2016. and East Asian plate tectonics since 52 Ma 3128 constrained by new subducted slab reconstruction methods. J. Geophys. Res. [Solid Earth] 121, 3129 4670–4741. 3130 Wu, L., Murphy, J.B., Quesada, C., Li, Z.-X., Waldron, J.W.F., Williams, S., Pisarevsky, S., Collins, 3131 W.J., 2020. The amalgamation of Pangea: Paleomagnetic and geological observations revisited. 3132 GSA Bulletin. https://doi.org/10.1130/B35633.1 3133 Xia, B., Zhang, L., Du, Z., Xu, B., 2017. Petrology and age of Precambrian Aksu blueschist, NW China. 3134 Precambrian Res. https://doi.org/10.1016/j.precamres.2017.12.041 3135 Xiao, W., Windley, B.F., Allen, M.B., Han, C., 2013. Paleozoic multiple accretionary and collisional 3136 tectonics of the Chinese Tianshan orogenic collage. Gondwana Res. 23, 1316–1341. 3137 https://doi.org/10.1016/j.gr.2012.01.012 3138 Xiao, W., Windley, B.F., Yong, Y., Yan, Z., Yuan, C., Liu, C., Li, J., 2009. Early Paleozoic to Devonian 3139 multiple-accretionary model for the Qilian Shan, NW China. J. Asian Earth Sci. 35, 323–333. 3140 https://doi.org/10.1016/j.jseaes.2008.10.001 3141 Xiao, W.J., Mao, Q.G., Windley, B.F., Han, C.M., Qu, J.F., Zhang, J.E., Ao, S.J., Guo, Q.Q., Cleven, 3142 N.R., Lin, S.F., Shan, Y.H., Li, J.L., 2010. Paleozoic multiple accretionary and collisional 3143 processes of the Beishan orogenic collage. Am. J. Sci. 310, 1553–1594. 3144 https://doi.org/10.2475/10.2010.12 3145 Xu, D., Xia, B., Bakun-Czubarow, N., Bachlinski, R., Li, P., Chen, G., Chen, T., 2008. Geochemistry and 3146 Sr-Nd isotope systematics of metabasites in the Tunchang area, Hainan Island, South China: 3147 implications for petrogenesis and tectonic setting. Mineral. Petrol. 92, 361–391. 3148 https://doi.org/10.1007/s00710-007-0198-0 3149 Xu, D.-R., Xia, B., Li, P.-C., Chen, G.-H., Ma, C., Zhang, Y.-Q., 2007. Protolith and U-Pb 3150 sensitive high mass-resolution ion microprobe (SHRIMP) zircon ages of the metabasites in 3151 Hainan Island, South China: Implications for geodynamic evolution since the late Precambrian. 3152 Island Arc 16, 575–597. 3153 Xu, X., Song, S., Su, L., Li, Z., Niu, Y., Allen, M.B., 2015. The 600–580Ma continental rift basalts in 3154 North Qilian Shan, northwest China: Links between the Qilian-Qaidam block and SE Australia, 3155 and the reconstruction of East Gondwana. Precambrian Res. 257, 47–64. 3156 https://doi.org/10.1016/j.precamres.2014.11.017 3157 Xu, Y., Cawood, P.A., Du, Y., Zhong, Z., Hughes, N.C., 2014. Terminal suturing of Gondwana along the 3158 southern margin of : Evidence from detrital zircon U-Pb ages and Hf isotopes 3159 in Cambrian and Ordovician strata, Hainan Island. Tectonics 33, 2490–2504. 3160 Yakubchuk, A., 2017. Evolution of the Central Asian Orogenic Supercollage since Late Neoproterozoic 3161 revised again. Gondwana Res. 47, 372–398. https://doi.org/10.1016/j.gr.2016.12.010 3162 Yang, B., Collins, A.S., Blades, M.L., 2019. Middle–late Mesoproterozoic tectonic geography of the 3163 North Australia Craton: U–Pb and Hf isotopes of detrital zircon grains in the Beetaloo Sub-basin, 3164 Northern …. Journal of the. 3165 Yang, B., Collins, A.S., Cox, G.M., Jarrett, A.J.M., Denyszyn, S., Blades, M.L., Farkaš, J., Glorie, S., 3166 2020. Using Mesoproterozoic sedimentary geochemistry to reconstruct basin tectonic geography 3167 and link organic carbon productivity to nutrient flux from a Northern Australian large igneous 3168 Province. Basin Res. 60, 175. https://doi.org/10.1111/bre.12450 3169 Yang, J.S., Wu, C.L., Chen, S.Y., Shi, R.D., Zhang, J.X., Meng, F.C., Zuo, G.C., Wu, H.Q., 3170 Constantinovskaya, E., 2006. Neoproterozoic eclogitic metamorphic age of the Beishan eclogite 3171 of Gansu, China: evidence from SHRIMP U--Pb isotope dating. Geology in China 33, 317–325. 3172 Yang, Z., Otofuji, Y.-I., Sun, Z., Huang, B., 2002. Magnetostratigraphic constraints on the Gondwanan 3173 origin of North China: Cambrian/Ordovician boundary results. Geophys. J. Int. 151, 1–10. 3174 https://doi.org/10.1046/j.1365-246X.2002.01656.x 3175 Yang, Z., Sun, Z., Yang, T., Pei, J., 2004. A long connection (750–380 Ma) between South China and 3176 Australia: paleomagnetic constraints. Earth Planet. Sci. Lett. 220, 423–434. 3177 https://doi.org/10.1016/S0012-821X(04)00053-6 3178 Yang, Z.-Y., Sun, Z.M., Ma, X.H., Huang, B.C., Dong, J.M., Zhou, Y.X., Zhu, H., 1996. Preliminary 3179 paleomagnetic results from the Lower Paleozoic of North China ( Province) and its 3180 implications. Chinese Science Bulletin (in Chinese) 42, 401. 3181 Yin, A., Harrison, T.M., 2000. Geologic Evolution of the Himalayan-Tibetan Orogen. Annu. Rev. Earth 3182 Planet. Sci. 28, 211–280. https://doi.org/10.1146/annurev.earth.28.1.211 3183 Young, A., Flament, N., Maloney, K., Williams, S., Matthews, K., Zahirovic, S., Müller, R.D., 2019. 3184 Global kinematics of tectonic plates and subduction zones since the late Paleozoic Era. 3185 Geoscience Frontiers 10, 989–1013. https://doi.org/10.1016/j.gsf.2018.05.011 3186 Yuan, C., Sun, M., Yang, J., Zhou, H., Zhou, M.-F., 2004. Nb-depleted, continental rift-related Akaz 3187 metavolcanic rocks (West Kunlun): implication for the rifting of the Tarim Craton from 3188 Gondwana. Geological Society, London, Special Publications 226, 131–143. 3189 https://doi.org/10.1144/GSL.SP.2004.226.01.07 3190 Zeng, Y.-C., Chen, Q., Xu, J.-F., Chen, J.-L., Huang, F., Yu, H.-X., Zhao, P.-P., 2018. Petrogenesis and 3191 geodynamic significance of Neoproterozoic ( 925 Ma) high-Fe–Ti of the RenTso

3192 ophiolite, Lhasa Terrane, central Tibet. Precambrian∼ Res. 314, 160–169. 3193 https://doi.org/10.1016/j.precamres.2018.06.005 3194 Zhan, S., Chen, Y., Xu, B., Wang, B., Faure, M., 2007. Late Neoproterozoic paleomagnetic results from 3195 the Sugetbrak Formation of the Aksu area, Tarim basin (NW China) and their implications to 3196 paleogeographic reconstructions and the snowball Earth hypothesis. Precambrian Res. 154, 143– 3197 158. https://doi.org/10.1016/j.precamres.2007.01.001 3198 Zhang, C.-L., Li, H.-K., Santosh, M., Li, Z.-X., Zou, H.-B., Wang, H., Ye, H., 2012. Precambrian 3199 evolution and cratonization of the Tarim Block, NW China: Petrology, geochemistry, Nd-isotopes 3200 and U–Pb zircon geochronology from Archaean -TTG–potassic granite suite and 3201 metamorphic belt. J. Asian Earth Sci. 47, 5–20. 3202 https://doi.org/10.1016/j.jseaes.2011.05.018 3203 Zhang, C.-L., Li, Z.-X., Li, X.-H., Ye, H.-M., 2009. Neoproterozoic mafic dyke swarms at the northern 3204 margin of the Tarim Block, NW China: Age, geochemistry, petrogenesis and tectonic 3205 implications. J. Asian Earth Sci. 35, 167–179. https://doi.org/10.1016/j.jseaes.2009.02.003 3206 Zhang, C.-L., Zou, H.-B., Li, H.-K., Wang, H.-Y., 2013. Tectonic framework and evolution of the Tarim 3207 Block in NW China. Gondwana Res. 23, 1306–1315. https://doi.org/10.1016/j.gr.2012.05.009 3208 Zhang, J., Mattinson, C.G., Meng, F., Wan, Y., Tung, K., 2008. Polyphase tectonothermal history 3209 recorded in granulitized gneisses from the north Qaidam HP/UHP metamorphic terrane, western 3210 China: Evidence from zircon U-Pb geochronology. Geol. Soc. Am. Bull. 120, 732–749. 3211 Zhang, N., Zhong, S., Leng, W., Li, Z.X., 2010. A model for the evolution of the Earth’s mantle structure 3212 since the Early Paleozoic. J. Geophys. Res. 3213 Zhang, Q.R., Piper, J.D.A., 1997. Palaeomagnetic study of Neoproterozoic glacial rocks of the Yangzi 3214 Block: palaeolatitude and configuration of South China in the late Proterozoic Supercontinent. 3215 Precambrian Res. 85, 173–199. https://doi.org/10.1016/S0301-9268(97)00031-4 3216 Zhang, S., Evans, D.A.D., Li, H., Wu, H., Jiang, G., Dong, J., Zhao, Q., Raub, T.D., Yang, T., 2013. 3217 Paleomagnetism of the late Cryogenian Nantuo Formation and paleogeographic implications for 3218 the South China Block. J. Asian Earth Sci. 72, 164–177. 3219 https://doi.org/10.1016/j.jseaes.2012.11.022 3220 Zhang, S., Li, H., Jiang, G., Evans, D.A.D., Dong, J., Wu, H., Yang, T., Liu, P., Xiao, Q., 2015. New 3221 paleomagnetic results from the Ediacaran in South China and their 3222 paleogeographic implications. Precambrian Res. 259, 130–142. 3223 https://doi.org/10.1016/j.precamres.2014.09.018 3224 Zhang, S.-H., Zhao, Y., Li, X.-H., Ernst, R.E., Yang, Z.-Y., 2017. The 1.33–1.30 Ga Yanliao large 3225 igneous province in the North China Craton: Implications for reconstruction of the Nuna 3226 (Columbia) supercontinent, and specifically with the North Australian Craton. Earth Planet. Sci. 3227 Lett. 465, 112–125. 3228 Zhao, G., Cawood, P.A., Wilde, S.A., Sun, M., 2002. Review of global 2.1–1.8 Ga orogens: implications 3229 for a pre-Rodinia supercontinent. Earth-Sci. Rev. 59, 125–162. https://doi.org/10.1016/S0012- 3230 8252(02)00073-9 3231 Zhao, G., Wang, Y., Huang, B., Dong, Y., Li, S., Zhang, G., Yu, S., 2018. Geological reconstructions of 3232 the East Asian blocks: From the breakup of Rodinia to the assembly of Pangea. Earth-Sci. Rev. 3233 186, 262–286. https://doi.org/10.1016/j.earscirev.2018.10.003 3234 Zhao, P., Chen, Y., Zhan, S., Xu, B., Faure, M., 2014. The Apparent Polar Wander Path of the Tarim 3235 block (NW China) since the Neoproterozoic and its implications for a long-term Tarim–Australia 3236 connection. Precambrian Res. 242, 39–57. https://doi.org/10.1016/j.precamres.2013.12.009 3237 Zhao, X., Coe, R.S., Liu, C., Zhou, Y., 1992. New Cambrian and Ordovician paleomagnetic poles for the 3238 North China Block and their paleogeographic implications. J. Geophys. Res., Studi. Geol. 97, 3239 1767–1788. https://doi.org/10.1029/91JB02742 3240 Zhong, S., Liu, X., 2016. The long-wavelength mantle structure and dynamics and implications for large- 3241 scale tectonics and volcanism in the Phanerozoic. Gondwana Res. 29, 83–104. 3242 https://doi.org/10.1016/j.gr.2015.07.007 3243 Zhong, S., Rudolph, M.L., 2015. On the temporal evolution of long-wavelength mantle structure of the 3244 Earth since the early Paleozoic. Geochem. Geophys. Geosyst. 16, 1599–1615. 3245 Zhong, S., Zhang, N., Li, Z.-X., Roberts, J.H., 2007. Supercontinent cycles, true polar wander, and very 3246 long-wavelength mantle convection. Earth Planet. Sci. Lett. 261, 551–564. 3247 https://doi.org/10.1016/j.epsl.2007.07.049 3248 Zhu, D.-C., Zhao, Z.-D., Niu, Y., Dilek, Y., Hou, Z.-Q., Mo, X.-X., 2013. The origin and pre-Cenozoic 3249 evolution of the Tibetan Plateau. Gondwana Res. 23, 1429–1454. 3250 https://doi.org/10.1016/j.gr.2012.02.002 3251 Zhu, D.-C., Zhao, Z.-D., Niu, Y., Dilek, Y., Mo, X.-X., 2011. Lhasa terrane in southern Tibet came from 3252 Australia. Geology 39, 727–730. https://doi.org/10.1130/G31895.1 3253 Zhu, G.-Y., Ren, R., Chen, F.-R., Li, T.-T., Chen, Y.-Q., 2017. Neoproterozoic rift basins and their 3254 control on the development of hydrocarbon source rocks in the Tarim Basin, NW China. J. Asian 3255 Earth Sci. 150, 63–72. https://doi.org/10.1016/j.jseaes.2017.09.018